Augmented reality-based hemiplegic hand intelligent evaluation and training method and device
By acquiring hand movement parameters of the hemiplegic hand, and combining them with the Brunnstrom and Fugl-Meyer scales for phased assessment, a fusion assessment result is generated. Training tasks are then matched and rehabilitation training is conducted using augmented reality devices. This solves the problems of single assessment results and subjective training tasks in existing technologies, and achieves multi-dimensional quantification of hemiplegic hand function assessment and sustainability of rehabilitation effects.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHENZHEN SECOND PEOPLES HOSPITAL (SHENZHEN INST OF TRANSLATIONAL MEDICINE)
- Filing Date
- 2026-04-16
- Publication Date
- 2026-07-14
AI Technical Summary
Existing methods for assessing and training hemiplegic hand function rely on manual scoring, which lacks objective quantification. This results in simplistic assessment outcomes and highly subjective selection of training tasks, making it difficult to guarantee the sustainability of rehabilitation effects.
By acquiring the hand movement parameters of the patient's hemiplegic hand, a phased assessment was conducted based on the Brunnstrom staging and Fugl-Meyer quantitative indicators to generate a fusion assessment result. Training tasks were then matched according to the assessment results, and rehabilitation training was carried out using augmented reality devices.
It enables multi-dimensional quantification of hemiplegic hand function assessment, improves assessment accuracy, reduces human intervention, ensures the continuity and effectiveness of the rehabilitation process, and supports home training and remote monitoring.
Smart Images
Figure CN122392800A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of data processing technology, and in particular to an intelligent assessment and training method and device for hemiplegic hand based on augmented reality. Background Technology
[0002] Rehabilitation of hemiplegic hand function is a core component of stroke sequelae treatment. Currently, assessment and training of hemiplegic hand function primarily involve doctors manually scoring the patient using a single clinical scale to determine the assessment results. Based on these results, therapists select training exercises for the patient based on their experience. The patient then follows the therapist's guidance to perform rehabilitation training for the hemiplegic hand.
[0003] This approach uses a single scale in the assessment phase, and the assessment results only reflect a single dimension of the patient's functional status, lacking a dual representation of the patient's functional stage and quantitative ability; in the training phase, there is no objective and quantitative correspondence between the selection of training tasks and the assessment results, relying on the therapist's subjective experience. Summary of the Invention
[0004] The main purpose of this application is to provide an augmented reality-based intelligent assessment and training method and device for hemiplegic hand, which aims to solve the technical problems of existing augmented reality-based intelligent assessment and training methods for hemiplegic hand being highly dependent on manual labor and difficult to ensure the continuity of rehabilitation effects in situations such as insufficient clinical manpower.
[0005] To achieve the above objectives, this application proposes an augmented reality-based intelligent assessment and training method for hemiplegic hands, which includes: Obtain hand movement parameters of the patient's hemiplegic hand; Based on the hand movement parameters, a phased assessment was performed to determine the patient's fusion assessment results; Based on the fusion assessment results, a training task for the hemiplegic hand is matched to the patient; The hemiplegic hand is given rehabilitation training based on the training task described above.
[0006] In one embodiment, the step of performing a phased assessment based on the hand movement parameters to determine the patient's fusion assessment result includes: Based on the hand movement parameters, Brunnstrom staging was performed to determine the Brunnstrom functional stage of the patient's hemiplegic hand. Obtain the Fugl-Meyer refined quantitative index corresponding to the Brunnstrom functional stage, and quantify and score the hand movement parameters based on the Fugl-Meyer refined quantitative index to determine the patient's Fugl-Meyer quantitative score within the Brunnstrom functional stage. The fusion assessment results of the patient were determined based on the Brunnstrom functional staging and the Fugl-Meyer quantitative score.
[0007] In one embodiment, the step of matching the patient with a training task for the hemiplegic hand based on the fusion assessment result includes: Determine the core action features corresponding to the Brunnstrom functional phases in the fusion evaluation results, and determine the initial difficulty parameters corresponding to the Fugl-Meyer quantitative score in the fusion evaluation results; The training task for the hemiplegic hand training is determined based on the core movement characteristics and the initial difficulty parameters.
[0008] In one embodiment, the step of acquiring the hand movement parameters of the hemiplegic hand of the patient includes: Spatial coordinates of key points in the patient's hemiplegic hand were collected using augmented reality devices. The hand movement parameters of the patient's hemiplegic hand are determined based on the spatial coordinates of the key points of the hand.
[0009] In one embodiment, the step of performing rehabilitation training on the hemiplegic hand based on the training task includes: Determine the guided movements for the hemiplegic hand corresponding to the training task; A virtual hand model is rendered in the display field of the augmented reality device, and the execution flow of the guided movements of the hemiplegic hand is demonstrated through the virtual hand model to guide the patient to perform rehabilitation training on the hemiplegic hand.
[0010] In one embodiment, the step of rendering a virtual hand model in the display field of the augmented reality device and demonstrating the execution flow of guided movements of the hemiplegic hand through the virtual hand model to guide the patient in rehabilitation training of the hemiplegic hand includes: Obtain the real position of the real image corresponding to the hemiplegic hand in the display field of view of the augmented reality device; In the display field of view, a virtual hand model is superimposed and rendered onto the real image of the hemiplegic hand according to the real position, and the execution process of the guided actions of the hemiplegic hand is demonstrated through the virtual hand model to guide the patient to carry out rehabilitation training for the hemiplegic hand.
[0011] In one embodiment, after the step of performing rehabilitation training on the hemiplegic hand based on the training task, the method further includes: The patient's hand movement parameters during the rehabilitation training are obtained, and the training task is updated based on the hand movement parameters during the rehabilitation training.
[0012] Furthermore, to achieve the above objectives, this application also proposes an augmented reality-based intelligent assessment and training device for hemiplegic hand, the augmented reality-based intelligent assessment and training device for hemiplegic hand comprising: The data acquisition module is used to acquire the hand movement parameters of the patient's hemiplegic hand; The hemiplegia assessment module is used to perform a phased assessment based on the hand movement parameters to determine the fusion assessment result of the patient. The task generation module is used to match the hemiplegic hand training task to the patient based on the fusion assessment results; The training guidance module is used to conduct rehabilitation training for the hemiplegic hand based on the training task.
[0013] Furthermore, to achieve the above objectives, this application also proposes an augmented reality-based intelligent assessment and training device for hemiplegic hand, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the augmented reality-based intelligent assessment and training method for hemiplegic hand as described above.
[0014] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the augmented reality-based intelligent assessment and training method for hemiplegic hand described above.
[0015] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the augmented reality-based intelligent assessment and training method for hemiplegic hand as described above.
[0016] One or more technical solutions proposed in this application have at least the following technical effects: This application obtains hand movement parameters of a patient's hemiplegic hand; performs phased assessments based on these parameters to determine the patient's fusion assessment result; matches the patient with a training task for the hemiplegic hand based on the fusion assessment result; and conducts rehabilitation training for the hemiplegic hand based on the training task. Because it generates a fusion assessment result containing both "qualitative staging and quantitative scoring" through phased assessments, it retains the practicality of the Brunnstrom staging system in clinical communication and training framework development, while also achieving the precise measurement of functional details using the Fugl-Meyer scale. This avoids the drawbacks of incomplete information from a single scale and the lack of logical connection between parallel scales, thus improving assessment accuracy. Furthermore, through automated assessment and rehabilitation training, it reduces the degree of human intervention in the hemiplegic hand rehabilitation process. Attached Figure Description
[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.
[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0019] Figure 1 This is a flowchart illustrating an embodiment of the augmented reality-based intelligent assessment and training method for hemiplegic hand in this application. Figure 2 This is a flowchart illustrating Embodiment 2 of the augmented reality-based intelligent assessment and training method for hemiplegic hand provided in this application. Figure 3 This is a flowchart illustrating Embodiment 3 of the augmented reality-based intelligent assessment and training method for hemiplegic hand provided in this application. Figure 4 This is a schematic diagram of the module structure of the augmented reality-based intelligent assessment and training device for hemiplegic hand according to an embodiment of this application; Figure 5 This is a schematic diagram of the hardware operating environment involved in the augmented reality-based intelligent assessment and training method for hemiplegic hand in this application embodiment.
[0020] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation
[0021] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.
[0022] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.
[0023] The main solution of this application is as follows: acquiring the hand movement parameters of the patient's hemiplegic hand; conducting phased assessments based on the hand movement parameters to determine the patient's fusion assessment results; matching training tasks for the hemiplegic hand based on the fusion assessment results; and conducting rehabilitation training for the hemiplegic hand based on the training tasks. This application, through automated data acquisition, phased assessment, and task matching based on assessment results, automates the core processes that traditionally rely on doctors' visual observation and subjective scoring, significantly reducing human intervention and subjective errors in the rehabilitation process. This provides a technological foundation for the home-based and large-scale application of hemiplegic hand rehabilitation. By conducting phased assessments and generating fusion assessment results, compared to existing technologies that rely on a single scale to obtain a single-dimensional score, it can simultaneously reflect the functional stage (qualitative dimension) and intra-stage quantitative ability of the patient's hemiplegic hand. This makes the assessment results a more comprehensive and accurate description of the patient's functional status, providing richer decision-making basis for subsequent training task matching.
[0024] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a computer or mobile phone, or an electronic device or virtual device capable of performing the above functions. The following description uses an augmented reality-based intelligent assessment and training device for hemiplegic hand (i.e., the patient's end) as an example to illustrate this embodiment and the following embodiments.
[0025] Based on this, embodiments of this application provide an intelligent assessment and training method for hemiplegic hand based on augmented reality, referring to... Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the augmented reality-based intelligent assessment and training method for hemiplegic hands provided in this application.
[0026] In this embodiment, the augmented reality-based intelligent assessment and training method for hemiplegic hand includes steps S10-S40: Step S10: Obtain the hand movement parameters of the patient's hemiplegic hand; Step S20: Based on the hand movement parameters, a phased assessment is performed to determine the fusion assessment result of the patient.
[0027] It should be noted that the augmented reality-based intelligent assessment and training method for hemiplegic hand in this application can be applied to the patient end of the augmented reality-based intelligent assessment and training system for hemiplegic hand. This patient end can upload rehabilitation data such as training data, assessment results, and quantitative indicators to the cloud server in real time, without requiring manual user operation, achieving full and traceable recording of home training data. The augmented reality-based intelligent assessment and training system for hemiplegic hand in this application can also include a doctor end. This doctor end allows doctors or therapists to remotely view all rehabilitation data uploaded to the cloud server by the patient end through a backend management system, review the Brunnstrom staging and Fugl-Meyer quantitative scores, assess rehabilitation progress, and manually adjust the training plan, difficulty, and virtual hand model guidance for problems encountered during patient training, achieving precise remote guidance. Through the augmented reality-based intelligent assessment and training method and system for hemiplegic hand in this application, patients can continue home rehabilitation training after discharge, receiving professional remote guidance and plan adjustments, achieving a complete sequential treatment of in-hospital assessment, home training, and remote monitoring, ensuring the continuity of rehabilitation effects and avoiding gaps in rehabilitation. Meanwhile, the cloud server in this application can perform statistical analysis on rehabilitation data from multiple patients, providing data support for optimizing clinical hemiplegic hand rehabilitation programs and conducting efficacy studies.
[0028] In some embodiments of this application, the patient terminal can be or can be connected to a lightweight augmented reality (AR) device. This AR device can integrate a front-facing infrared camera, a color camera, a visual display module, an audio module, and other devices or modules for input or output. Through these devices or modules, it is possible to capture the patient's full-joint hand movements, display virtual scenes / guidance / props, provide voice guidance and auditory feedback, etc., to facilitate the user's hemiplegia assessment and rehabilitation training.
[0029] In some embodiments of this application, the patient terminal may also be connected to a wireless wearable tactile module. The wireless wearable tactile module may integrate a micro vibration motor and a wireless communication module, and be connected to the patient terminal and / or augmented reality device to realize tactile feedback for the patient, with multi-level adjustable vibration intensity and frequency.
[0030] It should be noted that the cloud server can support the data storage, transmission, statistical analysis, and remote interaction of the augmented reality-based intelligent assessment and training system for hemiplegic hand, ensuring the realization of remote rehabilitation management functions.
[0031] In some embodiments of this application, the augmented reality device may include an AR hand motion capture module. This AR hand motion capture module enables the identification of key points and real-time acquisition and transmission of motion parameters of the fingertips, metacarpophalangeal joints, proximal interphalangeal joints, and distal interphalangeal joints, allowing the patient-side device to obtain the hand motion parameters of the patient's hemiplegic hand.
[0032] It should be noted that this application embodiment achieves a phased assessment of hand movement parameters based on the hemiplegic hand of the patient through dual-scale fusion. Specifically, this application uses the Brunnstrom scale stages 1-6 as a clinical grading framework, decomposes the refined quantitative indicators of the Fugl-Meyer scale according to the clinical function correspondence and embeds them into each stage, configuring a dedicated quantitative scoring standard for each stage; captures hand movement parameters of all joints of the hand, including the fingertips, metacarpophalangeal joints, proximal interphalangeal joints, and distal interphalangeal joints, through augmented reality devices; and uses a deep learning model to evaluate these hand movement parameters in stages, thereby determining the patient's fusion assessment result.
[0033] In this embodiment of the application, prior to the implementation of the above-mentioned scheme, the method further includes: constructing a Brunnstrom-Fugl-Meyer dual-scale fusion evaluation index system.
[0034] It should be noted that this application is based on the core fusion logic of "Brunnstrom framework and Fugl-Meyer decomposition and embedding," constructing a dual-scale fusion assessment index system and completing the design of a standardized training movement library. Specifically, it uses the Brunnstrom hemiplegic motor function assessment scale (stages 1-6) as the core clinical grading framework, retaining the core movement characteristics representing a qualitative leap in neurocontrol function in each stage; it decomposes the refined quantitative indicators of the Fugl-Meyer upper limb hand function assessment scale according to clinical function correspondence and embeds them one by one into each stage of Brunnstrom (stages 1-6); it develops a unique "core movement characteristics + Fugl-Meyer quantitative scoring standard" for each stage of Brunnstrom, clarifying the qualitative stage boundaries of each stage and achieving precise quantitative quantification of functional status within a single stage; based on this dual-scale fusion assessment system, it provides a standardized training movement library for Brunnstrom... Each of the six phases includes one basic movement and one task-oriented functional movement, forming a standardized training movement library. The basic movement aligns with the core movement characteristics within the phase, while the functional movement is adapted to daily life activities. At the same time, space is reserved for further refinement of movement design based on Fugl-Meyer quantitative indicators.
[0035] Understandably, the core motor characteristics of different stages of the Brunnstrom Hemiplegic Motor Function Assessment Scale can be: Stage 1: No active movement; Stage 2: Slight flexion; Stage 3: Active flexion; Stage 4: Dissociated movement; Stage 5: Fine motor function; Stage 6: Normal function.
[0036] It should be understood that by deconstructing and embedding Fugl-Meyer, it is possible to extract the scoring items related to hand function in the Fugl-Meyer scale one by one, and according to the clinical function correspondence, for each Brunnstrom stage, starting from the core motor characteristics of that stage, match the specific scoring items in Fugl-Meyer that can quantify the characteristic.
[0037] For example, in the simplified Fugl-Meyer motor function rating scale, the hand assessment items (Fugl-Meyer quantitative indicators) mainly include: group flexion: score 0-2 (0 = cannot flex, 1 = can flex but not sufficiently, 2 = can complete active flexion and extension); group extension: score 0-2 (0 = cannot extend, 1 = can relax actively flexed fingers, 2 = can fully actively extend); hook grip: score 0-2 (0 = cannot maintain, 1 = weak grip strength, 2 = can resist considerable resistance); lateral pinch: score 0-2 (0 = completely unable, 1 = weak pinch strength, 2 = can resist considerable resistance); cylindrical grip: score 0-2; spherical grip: score 0-2.
[0038] For example, in Brunnstrom stage 3, the core characteristic is "active flexion," specifically "the ability to perform full finger flexion and hooked grasping, but not extension." The corresponding Fugl-Meyer metric is that group flexion and hooked grasping are theoretically achievable.
[0039] It should be noted that by matching the Fugl-Meyer metric with the corresponding metric on the core action characteristics of a certain period after Brunnstrom's functional phases, the decomposed Fugl-Meyer metric is embedded into Brunnstrom's functional phases. This preserves the qualitative stage boundaries of each Brunnstrom period while achieving precise quantitative quantification of the functional status within a single period.
[0040] In some embodiments of this application, the embodiments can also design one basic movement and one task-oriented functional movement for each phase of Brunnstrom 1-6 based on the dual-scale fusion assessment system to form a standardized training movement library. The basic movement aligns with the core movement characteristics within the phase, the functional movement is adapted to daily life activities, and space is reserved for further refinement of movement design based on Fugl-Meyer quantitative indicators.
[0041] Understandably, by obtaining the fusion assessment results corresponding to the hand movement parameters, it is possible to determine which stage of the Brunnstrom functional stage the patient's hemiplegic hand is currently in, thus achieving a qualitative grading of the assessment. Within the determined Brunnstrom functional stage, a specific Fugl-Meyer score is determined, forming a precise assessment with clinical localization significance.
[0042] In some embodiments of this application, in order to achieve automatic evaluation, the deep learning model of this application can be trained based on Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM). This application does not limit its specific structure and training method, and can be designed according to the needs of actual applications.
[0043] In practice, the patient's device can acquire hand movement parameters of the hemiplegic hand; based on these parameters, a phased assessment is performed to determine the patient's fusion assessment result. Because the Brunnstrom functional stage of the hemiplegic hand and the Fugl-Meyer score within that stage are determined through phased assessment of hand movement parameters, a qualitative-to-quantitative phased assessment is achieved, improving assessment accuracy.
[0044] Step S30: Match the hemiplegic hand training task to the patient based on the fusion assessment results; Step S40: Perform recovery training on the hemiplegic hand based on the training task.
[0045] It is understood that the aforementioned training tasks may include training tasks related to basic movement training and training tasks related to functional movement training. Basic movement training is related to the core movement features corresponding to the Brunnstrom functional stage, while functional movement training is related to the Fugl-Meyer score within the Brunnstrom functional stage.
[0046] For example, for the core motor feature (dissociative movement) of Brunnstrom stage 4, it is necessary to establish the neural control ability of active finger extension as a basic motor training task. For Brunnstrom stage 4 with a Fugl-Meyer score of 5, virtual candy sharing can be designed as a functional movement to release the patient's grasping ability. The difficulty of this functional movement can be adaptively adjusted based on the Fugl-Meyer score.
[0047] In some embodiments of this application, after the step of performing rehabilitation training on the hemiplegic hand based on the training task, the method further includes: obtaining the patient's hand movement parameters in the rehabilitation training, and updating the training task based on the hand movement parameters in the rehabilitation training.
[0048] It is understood that the patient's hand movement parameters during each rehabilitation training session can be obtained through the patient's end, and then uploaded to the cloud server to update the training task. This update can be implemented through a motion server or a deep learning model designed in the patient's end, or it can be remotely updated through the doctor's backend management; this application embodiment does not impose any limitations on this.
[0049] This application embodiment acquires hand movement parameters of a patient's hemiplegic hand; performs phased assessments based on these parameters to determine the patient's fusion assessment result; matches the patient with a training task for the hemiplegic hand based on the fusion assessment result; and conducts rehabilitation training for the hemiplegic hand based on the training task. Because the phased assessment generates a fusion assessment result containing both "qualitative staging and quantitative scoring," it retains the practicality of the Brunnstrom staging system in clinical communication and training framework development, while also achieving the precise measurement of functional details using the Fugl-Meyer scale. This avoids the drawbacks of incomplete information from a single scale and the lack of logical connection between parallel scales, thus improving assessment accuracy. Through automated assessment and rehabilitation training, the degree of human intervention in the hemiplegic hand rehabilitation process is reduced.
[0050] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 , Figure 2 This is a flowchart illustrating Embodiment 2 of the augmented reality-based intelligent assessment and training method for hemiplegic hand provided in this application.
[0051] like Figure 2 As shown in this embodiment, the step of determining the patient's fusion assessment result based on the hand movement parameters through a phased assessment includes: Step S21: Based on the hand movement parameters, Brunnstrom staging is determined to obtain the Brunnstrom functional stage to which the patient's hemiplegic hand belongs; Step S22: Obtain the Fugl-Meyer refined quantitative index corresponding to the Brunnstrom functional stage, and quantify and score the hand movement parameters based on the Fugl-Meyer refined quantitative index to determine the patient's Fugl-Meyer quantitative score within the Brunnstrom functional stage. Step S23: Determine the fusion assessment result of the patient based on the Brunnstrom functional staging and the Fugl-Meyer quantitative score.
[0052] It should be noted that, in order to achieve AI intelligent assessment and training task matching, the embodiments of this application can input the hand motion parameters captured by the augmented reality device into a deep learning model that integrates CNN and LSTM. In this model, CNN is responsible for extracting the spatial features of the hand motion parameters (such as the angles and positions of each joint), and LSTM is responsible for extracting the temporal series features of the hand motion parameters (such as the time of action completion and the trajectory of the movement). The deep learning model strictly follows the dual-scale fusion logic of "qualitative first, quantitative later" to complete the accurate matching of intelligent assessment and training tasks.
[0053] In some embodiments of this application, based on the fusion assessment results, a report including Brunnstrom functional staging, corresponding Fugl-Meyer quantitative scores, and standardized assessment of core motor features can also be output. The assessment results can be manually assessed by clinical rehabilitation physicians as the gold standard, with a consistency of ≥85%.
[0054] Understandably, the Brunnstrom staging system is used to perform qualitative analysis, and to accurately match hand movement parameters with the core movement characteristics of Brunnstrom stages 1-6 in the dual-scale fusion assessment index system.
[0055] For example, in this embodiment, spatial features extracted by CNN and temporal features extracted by LSTM are concatenated. The spatial features are then processed using a staging classification head (which can consist of a fully connected layer and a softmax classifier) to obtain the probabilities corresponding to different Brunnstrom functional stages. The Brunnstrom functional stage with the highest probability is taken as the Brunnstrom functional stage to which the patient's hemiplegic hand belongs. Further, a scoring prediction head corresponding to the Brunnstrom functional stage can be selected. This scoring prediction head can be multiple scoring prediction heads set in parallel, each corresponding to one item in the Fugl-Meyer hand assessment. By accumulating the item scores from multiple scoring prediction heads, the patient's Fugl-Meyer quantitative score within the Brunnstrom functional stage can be obtained.
[0056] In some embodiments of this application, the step of matching the training task of the hemiplegic hand to the patient based on the fusion assessment result includes: determining the core motor features corresponding to the Brunnstrom functional stage in the fusion assessment result, and determining the initial difficulty parameter corresponding to the Fugl-Meyer quantitative score in the fusion assessment result; and determining the training task of the hemiplegic hand training based on the core motor features and the initial difficulty parameter.
[0057] It should be noted that the aforementioned initial difficulty parameters can be parameters used to control the training difficulty of functional movements, determined based on the Fugl-Meyer quantitative score, such as controlling the percentage increase / decrease in the size of AR virtual props, controlling the percentage increase / decrease in virtual weight, completion time limits, etc. This application embodiment does not impose any limitations on these parameters. A higher Fugl-Meyer quantitative score corresponds to a greater initial difficulty of the functional movement, and a lower Fugl-Meyer quantitative score corresponds to a lower initial difficulty. Simultaneously, this application embodiment can provide evaluation feedback based on the user's hand movement parameters after training, using the Fugl-Meyer quantitative score to advance the gradient upgrade of the training task, achieving a complete training-evaluation closed loop of "evaluation-training-re-evaluation-retraining".
[0058] It should be explained that the above-mentioned functional movements are daily life movements evolved from basic movements, with the goal of improving the rehabilitation effect and transferring it to real life, such as holding a cup, pinching beans, grabbing keys, wringing a towel, and buttoning buttons. This application embodiment does not limit these.
[0059] For example, in Brunnstrom Stage 3, the core feature of this stage is "active flexion," specifically "the ability to flex all fingers and perform a hook-like grasp, but not to extend them." The corresponding basic movement training could be hook-like grasp training, and the functional movement training could be lifting a virtual shopping bag to a virtual car in the AR field of view. The initial difficulty parameters could be used to control the weight of the virtual shopping bag, the completion time limit, etc.
[0060] This application embodiment determines the Brunnstrom functional stage of the patient's hemiplegic hand by determining the Brunnstrom stage based on hand movement parameters; obtains the Fugl-Meyer refined quantitative index corresponding to the Brunnstrom functional stage, and quantifies and scores the hand movement parameters based on the Fugl-Meyer refined quantitative index to determine the patient's Fugl-Meyer quantitative score within the Brunnstrom functional stage; and determines the patient's fusion assessment result based on the Brunnstrom functional stage and Fugl-Meyer quantitative score. Because it is based on the organic fusion of two scales—"Brunnstrom framework, Fugl-Meyer deconstruction and embedding"—and combined with AR real-time full-joint fine motion capture and AI deep learning algorithms, it constructs a complete hemiplegic hand rehabilitation technology solution: "dual-scale fusion assessment → AR fine motion capture → AI intelligent assessment matching → tiered training → multimodal guided feedback → quantitative evaluation → remote management". The core logic is as follows: using the Brunnstrom scale stages 1-6 as a clinical grading framework, the refined quantitative indicators of the Fugl-Meyer scale are decomposed according to the clinical functional correspondence and embedded into each stage, and a unique quantitative scoring standard is configured for each stage; AR glasses are used to capture fine motion parameters of all joints of the hand; a CNN+LSTM deep learning model is used to complete the AI intelligent assessment of "first determining the Brunnstrom stage, then calculating the Fugl-Meyer score within the stage", as well as the precise matching with training tasks; a visual AR guidance program without speech dependence is designed for patients with hearing impairment and aphasia, and a unique guidance format is designed to match the type of training task; multimodal feedback and gamified incentives are integrated to improve training compliance; a whole-joint quantitative evaluation index system is established; and finally, a remote rehabilitation management system is used to realize home training and sequential treatment, solving the problems of insufficient clinical manpower and gaps in home rehabilitation, and realizing the precision, personalization, functionality, intelligence and continuity of hemiplegic hand rehabilitation.
[0061] Based on the first and / or second embodiments of this application, in the third embodiment of this application, the content that is the same as or similar to the first and / or second embodiments described above can be referred to the above description and will not be repeated hereafter. Based on this, please refer to... Figure 3 , Figure 3 This is a flowchart illustrating Embodiment 3 of the augmented reality-based intelligent assessment and training method for hemiplegic hand provided in this application.
[0062] like Figure 3 As shown in the embodiment of this application, the step of obtaining the hand movement parameters of the hemiplegic hand of the patient includes: Step S11: Collect the spatial coordinates of key points of the patient's hemiplegic hand using augmented reality equipment; Step S12: Determine the hand movement parameters of the patient's hemiplegic hand based on the spatial coordinates of the key hand points.
[0063] It should be noted that the aforementioned hand key point spatial coordinates refer to the three-dimensional spatial coordinates of the hand captured by identifying and tracking four core key points of the patient's hemiplegic hand: fingertips, metacarpophalangeal joints (MCP), proximal interphalangeal joints (PIP), and distal interphalangeal joints (DIP). In this embodiment, lightweight AR glasses equipped with a front-facing infrared camera and a color camera can be used to identify and track the four core key points of the patient's hemiplegic hand, capturing the three-dimensional spatial motion parameters of the hand in real time. Specifically, these parameters include: flexion / extension angles of each joint, thumb opposition angle, metacarpophalangeal joint range of motion, grasp / release amplitude, action completion time, and movement sway amplitude (stability). The camera sampling frequency can be ≥30fps, and the hand key point recognition accuracy rate is ≥90%. The captured hand motion parameters are transmitted in real time to the AI intelligent analysis module and simultaneously synchronized to the remote rehabilitation management system.
[0064] In some embodiments of this application, in order to improve the training effect, the step of rendering a virtual hand model in the display field of view of the augmented reality device and demonstrating the execution flow of the hemiplegic hand guided actions through the virtual hand model to guide the patient to perform rehabilitation training on the hemiplegic hand includes: obtaining the real position of the real image corresponding to the hemiplegic hand in the display field of view of the augmented reality device; in the display field of view, superimposing and rendering the virtual hand model onto the real image of the hemiplegic hand according to the real position, and demonstrating the execution flow of the hemiplegic hand guided actions through the virtual hand model to guide the patient to perform rehabilitation training on the hemiplegic hand.
[0065] It should be noted that the embodiments of this application use AR multi-form virtual guidance to adapt to various types of hemiplegic patients, and have designed a visual core guidance scheme without verbal dependence. At the same time, it is combined with voice commands (optionally turned off) to achieve full coverage of patients in all functional states; and matches exclusive guidance forms according to the training task type (basic movements / task-oriented functional movements) to improve guidance accuracy. Specific guidance forms and suitable scenarios are also discussed.
[0066] In some embodiments of this application, the embodiments of this application can demonstrate and guide independently using a virtual hand model. An independent virtual hand model is generated in the display field of the AR device to demonstrate the complete process of training actions purely visually. At the same time, voice commands can be selectively added to explain the key points of the actions.
[0067] In some embodiments of this application, the present application can demonstrate and guide by superimposing a virtual hand model and a real hand. In the display field of the AR device, a semi-transparent standard virtual hand is superimposed on the real position of the patient's hemiplegic hand to achieve embodied imitation guidance and exclusive adaptation of basic movement training. This allows the patient to intuitively see the difference between the standard movement and the movement of their own joints, which is used for accurate movement calibration, correction of compensatory movements, and improvement of the quality of basic movement completion.
[0068] In some embodiments of this application, demonstration and guidance can be provided by combining a real hand with virtual assistive devices. This demonstration and guidance can be specifically adapted to functional movement training. Virtual assistive devices (such as virtual cups, balls, keys, etc.) that match the training task are superimposed on the real image of the patient's hemiplegic hand. The size, weight, and other parameters of the virtual props can be adjusted arbitrarily according to the initial training difficulty to achieve precise quantification of the training content. Moreover, there are no safety hazards associated with physical props, and training conditions that cannot be achieved in real scenarios, such as shock absorption, can be simulated. This allows patients to complete actions with real-life significance, such as grasping, pinching, and manipulating, achieving precise connection between training actions and daily life activities, and improving the ability to transfer rehabilitation effects to daily life.
[0069] In some embodiments of this application, the embodiments of this application may employ any one or more of the above-described demonstration guidance methods to achieve training demonstration guidance.
[0070] In some embodiments of this application, during the training process, a multimodal feedback mechanism integrating visual, auditory, and tactile feedback can be integrated, along with a gamified incentive mechanism to improve patient training compliance. Specific feedback forms are as follows: 1. Visual feedback: After the patient completes a standard action, virtual props trigger color-changing, glowing, and explosion effects. The AR field of view displays a training progress bar and reward markers such as stars / badges in real time, providing intuitive feedback on the quality of action completion; 2. Auditory feedback: After the patient completes a standard action, optional action completion prompts and periodic encouraging voices can be triggered. Rhythmic background sound effects can be optionally added during training; 3. Tactile feedback: This is achieved through a wearable micro-vibration module wirelessly connected to AR glasses. The degree of action compliance is positively correlated with vibration intensity and frequency. The more standard the action, the clearer and more regular the vibration, achieving precise tactile feedback; 4. Gamified incentive mechanism: Patients can earn corresponding points by completing training tasks. These points can be redeemed for virtual training scenes, virtual props, and training levels, enabling gamified progression and enhancing the fun and sustainability of training.
[0071] In some embodiments of this application, the hand movement data and core quantitative evaluation indicators during the rehabilitation training process can also be used as the core basis for judging rehabilitation progress and adjusting training difficulty. The evaluation results are synchronized to the remote rehabilitation management system (i.e., cloud server) for doctors to view remotely. Specific indicators can be as follows: 1. Angle indicators: flexion / extension range and joint mobility of each joint (fingertip, metacarpophalangeal, proximal interphalangeal, distal interphalangeal) of the hemiplegic hand, corresponding to the Fugl-Meyer joint mobility quantitative indicators; 2. Stability indicators: the swaying amplitude of hand movements, the coordination of each joint movement, and the compensation situation when completing the movement, corresponding to the Fugl-Meyer movement quality quantitative indicators; 3. Time indicators: the time required to complete a single standard movement. The shorter the time, the better the motor control ability, corresponding to the Fugl-Meyer movement completion efficiency quantitative indicators; 4. Repetition / endurance indicators: the number of times the standard movement can be completed continuously in a single training session. The more repetitions, the better the exercise endurance, serving as a supplement to the Fugl-Meyer quantitative indicators to assess the ability to sustain movement.
[0072] This application's embodiments use AR devices to collect the spatial coordinates of key points on the hemiplegic hand of a patient; based on these spatial coordinates, the hand movement parameters of the hemiplegic hand are determined. This application pioneers a dual-scale fusion assessment system that uses "Brunnstrom for framing and Fugl-Meyer decomposition and embedding," decomposing Fugl-Meyer quantitative indicators according to their clinical functional correspondence and embedding them into Brunnstrom. Each of the six phases is equipped with a dedicated quantitative scoring standard, retaining the clinical practicality of Brunnstrom while achieving the quantitative accuracy of Fugl-Meyer. The assessment results have a consistency of ≥85% with clinical expert manual assessments, and the matching degree with training tasks is significantly improved. By optimizing the recognition range of key hand points, full joint capture of fingertips, metacarpophalangeal joints, proximal interphalangeal joints, and distal interphalangeal joints can be achieved, accurately acquiring core motion parameters such as flexion angle and stability of each joint, especially meeting the fine motor rehabilitation needs of hemiplegic hands, with a motion capture accuracy rate of ≥90%. By designing a non-verbal dependent visual AR guidance scheme adapted to patients with hearing impairment and aphasia, and designing multiple forms of guidance to match the training task type, the needs of all patients are taken into account, improving training cooperation and compliance. Based on the dual-scale fusion logic, the "first determine Brunnstrom" is completed. The system employs intelligent assessment of Fugl-Meyer scores during the Unnstrom and recalculation phases, along with precise matching of assessment results with training tasks and adaptive difficulty adjustments, forming a complete intelligent closed loop of "assessment-training-feedback." This significantly reduces reliance on manual intervention and alleviates the shortage of clinical rehabilitation personnel. By utilizing the advantages of virtual props—such as the ability to flexibly adjust parameters like size and weight—training content is precisely quantified. A tiered training system combining basic movements and task-oriented functional movements is designed, closely integrating training with daily living activities to ensure the transfer of rehabilitation effects to real life. This achieves automation, lightweighting, and home-based assessment and training, allowing patients to independently complete home training. A supporting remote rehabilitation management system enables real-time data transmission, remote monitoring by doctors, and plan adjustments, ensuring sequential treatment after discharge and completely resolving the gap in home rehabilitation.
[0073] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the intelligent assessment and training method for hemiplegic hand based on augmented reality. Any simple modifications based on this technical concept are within the scope of protection of this application.
[0074] This application also provides an intelligent assessment and training device for hemiplegic hand based on augmented reality, please refer to... Figure 4 , Figure 4 This is a schematic diagram of the module structure of the augmented reality-based intelligent assessment and training device for hemiplegic hand according to an embodiment of this application. The augmented reality-based intelligent assessment and training device for hemiplegic hand includes: Data acquisition module 10 is used to acquire hand movement parameters of the patient's hemiplegic hand; The hemiplegia assessment module 20 is used to perform a phased assessment based on the hand movement parameters to determine the fusion assessment result of the patient. The task generation module 30 is used to match the hemiplegic hand training task to the patient based on the fusion assessment results. The training guidance module 40 is used to conduct rehabilitation training for the hemiplegic hand based on the training task.
[0075] The augmented reality-based intelligent assessment and training device for hemiplegic hand provided in this application, employing the augmented reality-based intelligent assessment and training method for hemiplegic hand in the above embodiments, can solve the technical problems of existing augmented reality-based intelligent assessment and training methods for hemiplegic hand being highly dependent on manual labor and difficult to guarantee the continuity of rehabilitation effects in situations such as insufficient clinical manpower. Compared with the prior art, the beneficial effects of the augmented reality-based intelligent assessment and training device for hemiplegic hand provided in this application are the same as the beneficial effects of the augmented reality-based intelligent assessment and training method for hemiplegic hand provided in the above embodiments, and other technical features in the augmented reality-based intelligent assessment and training device for hemiplegic hand are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.
[0076] This application provides an augmented reality-based intelligent assessment and training device for hemiplegic hand. The augmented reality-based intelligent assessment and training device for hemiplegic hand includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the augmented reality-based intelligent assessment and training method for hemiplegic hand in the above embodiment 1.
[0077] The following is for reference. Figure 5 The diagram illustrates a structural schematic of an augmented reality-based intelligent assessment and training device for hemiplegic hands suitable for implementing embodiments of this application. The augmented reality-based intelligent assessment and training device for hemiplegic hands in embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 5 The augmented reality-based intelligent assessment and training device for hemiplegic hand shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.
[0078] like Figure 5 As shown, the augmented reality-based intelligent assessment and training device for hemiplegic hand may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to programs stored in read-only memory (ROM) 1002 or programs loaded from storage device 1003 into random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the augmented reality-based intelligent assessment and training device for hemiplegic hand. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the augmented reality-based intelligent assessment and training device for hemiplegic hands to exchange data wirelessly or via wired communication with other devices. Although the figure shows an augmented reality-based intelligent assessment and training device for hemiplegic hands with various systems, it should be understood that implementing or having all the systems shown is not required. More or fewer systems can be implemented alternatively.
[0079] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.
[0080] The augmented reality-based intelligent assessment and training device for hemiplegic hand provided in this application, employing the augmented reality-based intelligent assessment and training method for hemiplegic hand in the above embodiments, can solve the technical problem that existing augmented reality-based intelligent assessment and training methods for hemiplegic hand are highly dependent on manual labor and cannot guarantee the continuity of rehabilitation effects in situations such as insufficient clinical manpower. Compared with the prior art, the beneficial effects of the augmented reality-based intelligent assessment and training device for hemiplegic hand provided in this application are the same as those of the augmented reality-based intelligent assessment and training method for hemiplegic hand provided in the above embodiments, and other technical features in this augmented reality-based intelligent assessment and training device for hemiplegic hand are the same as those disclosed in the previous embodiment method, and will not be repeated here.
[0081] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.
[0082] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
[0083] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the augmented reality-based intelligent assessment and training method for hemiplegic hand described in the above embodiments.
[0084] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or flash memory, optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.
[0085] The aforementioned computer-readable storage medium may be included in an augmented reality-based intelligent assessment and training device for hemiplegic hands; or it may exist independently and not incorporated into the augmented reality-based intelligent assessment and training device for hemiplegic hands.
[0086] The aforementioned computer-readable storage medium carries one or more programs that, when executed by the augmented reality-based intelligent assessment and training device for hemiplegic hand, cause the augmented reality-based intelligent assessment and training device for hemiplegic hand to: Obtain hand movement parameters of the patient's hemiplegic hand; Based on the hand movement parameters, a phased assessment was performed to determine the patient's fusion assessment results; Based on the fusion assessment results, a training task for the hemiplegic hand is matched to the patient; The hemiplegic hand is given rehabilitation training based on the training task described above.
[0087] Computer program code for performing the operations of this application can be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, and C++, as well as conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).
[0088] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation that may be implemented in systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.
[0089] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.
[0090] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., computer programs) for executing the aforementioned augmented reality-based intelligent assessment and training method for hemiplegic hands. This addresses the technical problem that existing augmented reality-based intelligent assessment and training methods for hemiplegic hands are heavily reliant on manual labor, making it difficult to guarantee the continuity of rehabilitation effects in situations such as insufficient clinical manpower. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the augmented reality-based intelligent assessment and training method for hemiplegic hands provided in the above embodiments, and will not be repeated here.
[0091] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the augmented reality-based intelligent assessment and training method for hemiplegic hand as described above.
[0092] The computer program product provided in this application can solve the technical problem that existing augmented reality-based intelligent assessment and training methods for hemiplegic hands are highly dependent on manual labor and cannot guarantee the continuity of rehabilitation effects when there is a shortage of clinical staff. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the augmented reality-based intelligent assessment and training method for hemiplegic hands provided in the above embodiments, and will not be repeated here.
[0093] The above description is only a part of the embodiments of this application and does not limit the scope of protection of this application. All equivalent structural transformations made under the technical concept of this application and using the content of this application specification and drawings, or direct / indirect applications in other related technical fields, are included in the scope of protection of this application.
Claims
1. An intelligent assessment and training method for hemiplegic hand based on augmented reality, characterized in that, The method includes: Obtain hand movement parameters of the patient's hemiplegic hand; Based on the hand movement parameters, a phased assessment was performed to determine the patient's fusion assessment results; Based on the fusion assessment results, a training task for the hemiplegic hand is matched to the patient; The hemiplegic hand is given rehabilitation training based on the training task described above.
2. The augmented reality-based intelligent assessment and training method for hemiplegic hand as described in claim 1, characterized in that, The step of performing a phased assessment based on the hand movement parameters to determine the patient's fusion assessment result includes: Based on the hand movement parameters, Brunnstrom staging was performed to determine the Brunnstrom functional stage of the patient's hemiplegic hand. Obtain the Fugl-Meyer refined quantitative index corresponding to the Brunnstrom functional stage, and quantify and score the hand movement parameters based on the Fugl-Meyer refined quantitative index to determine the patient's Fugl-Meyer quantitative score within the Brunnstrom functional stage. The fusion assessment results of the patient were determined based on the Brunnstrom functional staging and the Fugl-Meyer quantitative score.
3. The augmented reality-based intelligent assessment and training method for hemiplegic hand as described in claim 2, characterized in that, The step of matching the hemiplegic hand training task to the patient based on the fusion assessment results includes: Determine the core action features corresponding to the Brunnstrom functional phases in the fusion evaluation results, and determine the initial difficulty parameters corresponding to the Fugl-Meyer quantitative score in the fusion evaluation results; The training task for the hemiplegic hand training is determined based on the core movement characteristics and the initial difficulty parameters.
4. The augmented reality-based intelligent assessment and training method for hemiplegic hand as described in claim 1, characterized in that, The steps for obtaining the hand movement parameters of the patient's hemiplegic hand include: Spatial coordinates of key points in the patient's hemiplegic hand were collected using augmented reality devices. The hand movement parameters of the patient's hemiplegic hand are determined based on the spatial coordinates of the key points of the hand.
5. The augmented reality-based intelligent assessment and training method for hemiplegic hand as described in claim 4, characterized in that, The steps for performing rehabilitation training on the hemiplegic hand based on the training task include: Determine the guided movements for the hemiplegic hand corresponding to the training task; A virtual hand model is rendered in the display field of the augmented reality device, and the execution flow of the guided movements of the hemiplegic hand is demonstrated through the virtual hand model to guide the patient to perform rehabilitation training on the hemiplegic hand.
6. The augmented reality-based intelligent assessment and training method for hemiplegic hand as described in claim 5, characterized in that, The steps of rendering a virtual hand model in the display field of the augmented reality device and demonstrating the execution flow of guided movements of the hemiplegic hand through the virtual hand model to guide the patient in rehabilitation training of the hemiplegic hand include: Obtain the real position of the real image corresponding to the hemiplegic hand in the display field of view of the augmented reality device; In the display field of view, a virtual hand model is superimposed and rendered onto the real image of the hemiplegic hand according to the real position, and the execution process of the guided actions of the hemiplegic hand is demonstrated through the virtual hand model to guide the patient to carry out rehabilitation training for the hemiplegic hand.
7. The augmented reality-based intelligent assessment and training method for hemiplegic hand as described in any one of claims 1-6, characterized in that, Following the step of performing rehabilitation training on the hemiplegic hand based on the training task, the method further includes: The patient's hand movement parameters during the rehabilitation training are obtained, and the training task is updated based on the hand movement parameters during the rehabilitation training.
8. An intelligent assessment and training device for hemiplegic hand based on augmented reality, characterized in that, The augmented reality-based intelligent assessment and training device for hemiplegic hand includes: The data acquisition module is used to acquire the hand movement parameters of the patient's hemiplegic hand; The hemiplegia assessment module is used to perform a phased assessment based on the hand movement parameters to determine the fusion assessment result of the patient. The task generation module is used to match the hemiplegic hand training task to the patient based on the fusion assessment results; The training guidance module is used to conduct rehabilitation training for the hemiplegic hand based on the training task.
9. An intelligent assessment and training device for hemiplegic hand based on augmented reality, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the augmented reality-based intelligent assessment and training method for hemiplegic hand as described in any one of claims 1 to 7.
10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the augmented reality-based intelligent assessment and training method for hemiplegic hand as described in any one of claims 1 to 7.