Cerebral stroke dysphagia rehabilitation training interaction method and system

Abstract

The invention provides a stroke dysphagia rehabilitation training interaction method and system, and relates to the technical field of wisdom medical treatment, and the method constructs a closed-loop and multi-dimensional rehabilitation interaction system through integrating four modules of swallowing function evaluation, course execution, psychological state monitoring and AR gamification training. A patient can complete structured evaluation of swallowing ability and a family support environment on a main interface in a one-stop mode, an adjustable personalized rehabilitation path is generated accordingly, and feedback is dynamically collected and executed in the training process to optimize course content; meanwhile, the emotion evaluation mechanism can timely identify negative psychological states and automatically trigger regulation intervention; by combining the augmented reality technology, food or tableware is recognized in real time in a real feeding scene, risk prompt and safety guidance are overlaid, the training interestingness and compliance are improved through interactive games, finally individuation, visualization, emotional and immersive fusion of the rehabilitation process is achieved, and the rehabilitation effect and life quality of stroke dysphagia patients are improved.

Description

Technical Field

[0001] This application relates to the field of smart medical technology, and in particular to an interactive method and system for rehabilitation training of dysphagia after stroke. Background Technology

[0002] Stroke patients often experience swallowing difficulties, which not only affect nutritional intake and rehabilitation progress but can also lead to serious complications such as aspiration pneumonia. Therefore, systematic and personalized rehabilitation interventions are urgently needed. Current technologies rely heavily on clinical scales completed by professionals for swallowing function assessment, lacking patient participation mechanisms, and the assessment results are difficult to dynamically guide subsequent training. Rehabilitation training content is usually fixed and lacks engaging design, resulting in low patient compliance and difficulty in long-term adherence. Psychological state has a significant impact on rehabilitation outcomes, but current programs generally neglect the integration of emotion monitoring and regulation. Furthermore, while the family support environment is recognized as an important factor influencing rehabilitation effectiveness, it is rarely systematically incorporated into the rehabilitation pathway generation logic. Although some programs attempt to introduce remote monitoring technology, it is often functionally isolated, failing to achieve a closed-loop linkage between assessment, training, psychological intervention, medical care collaboration, and family participation. This leads to a fragmented rehabilitation process, delayed feedback, insufficient personalization, and limited overall rehabilitation efficiency. Summary of the Invention

[0003] This application provides an interactive method and system for rehabilitation training of dysphagia in stroke patients, in order to solve one or more technical problems existing in the prior art, and at least provide a beneficial option or create conditions that can realize a closed-loop linkage of assessment, training, psychological intervention, medical care collaboration and family participation, thereby improving patient compliance and rehabilitation outcomes.

[0004] On the one hand, this application provides an interactive method for rehabilitation training of dysphagia in stroke patients, including the following steps: The main interface of the patient terminal for stroke dysphagia rehabilitation training is displayed. The main interface includes swallowing function assessment controls, course execution controls, psychological assessment controls, and AR gamified training controls. In response to the triggering operation of the swallowing function assessment control, a swallowing function assessment sub-interface is displayed, structured data of the patient in the swallowing function dimension and family support dimension are collected, a swallowing function assessment report is generated and a set of rehabilitation goals is determined, the patient's confirmation or adjustment instructions for the set of rehabilitation goals are received, and a rehabilitation path is generated; the rehabilitation path includes a training course sequence and its difficulty parameters. In response to the triggering operation of the course execution control, the course execution sub-interface is displayed, the training course corresponding to the rehabilitation path is loaded, the execution feedback and obstacle description submitted by the patient are obtained after the training is completed, and the training course is adjusted based on the execution feedback and obstacle description. In response to the triggering operation of the psychological assessment control, the psychological assessment sub-interface is displayed, and an emotion assessment is performed. When an assessment result that matches a preset negative emotion pattern is detected, the emotion regulation mode is triggered. In response to the triggering operation of the AR gamified training control, an AR training sub-interface is displayed, which identifies food or tableware in the field of vision in real time, displays swallowing risk warnings and safe eating guidelines, loads a swallowing interactive game, provides feedback based on the execution result of the swallowing action, and updates achievement marks and incentive information.

[0005] Furthermore, the swallowing function assessment sub-interface includes swallowing function scale controls and family support questionnaire controls; In the swallowing function assessment sub-interface, structured data of the patient in the swallowing function dimension and family support dimension are collected, a swallowing function assessment report is generated, and a set of rehabilitation goals is determined. The system receives confirmation or adjustment instructions from the patient regarding the set of rehabilitation goals and generates a rehabilitation path, including the following steps: In response to a trigger operation on the swallowing function scale control, the swallowing function assessment scale is loaded and displayed, and the patient's score input for each swallowing ability item is received as swallowing function information. In response to the triggering operation of the family support questionnaire control, the structured questions of the home rehabilitation environment are loaded and displayed, and inputs about caregiver configuration and home training conditions are received as family support information; Based on the swallowing function information and the family support information, a swallowing function assessment report is generated, and a set of rehabilitation goals is constructed accordingly. The set of rehabilitation goals is displayed in the swallowing function assessment sub-interface, which receives the patient's confirmation or adjustment instructions for the rehabilitation goals and generates the rehabilitation path based on the confirmed set of rehabilitation goals.

[0006] Furthermore, the course execution sub-interface includes a course content display area, a training check-in control, and a feedback input control; In the course execution sub-interface, the training course corresponding to the rehabilitation path is loaded. After the training is completed, the execution feedback and obstacle description submitted by the patient are obtained, and the training course is adjusted based on the execution feedback and obstacle description, including the following steps: In the course content display area, training courses corresponding to the rehabilitation path are loaded and displayed, presenting the key points of swallowing training movements in the form of videos or text. In response to the triggering operation of the training check-in control, the completion status of the current course is recorded, and subsequent operations are locked until feedback is submitted; In response to the triggering operation of the feedback input control, a training feedback window is displayed to receive selective input from the patient regarding the training execution status. The selective input includes the degree of action completion, the type of subjective difficulty, and a description of the body's reaction. The feedback data is then associated with the current course record, and the content of the training course is adjusted accordingly.

[0007] Furthermore, the step of associating the feedback data with the current course record and adjusting the content of the training course accordingly specifically includes: Using a pre-defined rehabilitation progress prediction model, the model takes the patient's historical training completion rate, the frequency of the distribution of obstacle types, and the trend of changes in emotional scores as input features, and outputs the patient's rehabilitation stagnation probability. When the probability of rehabilitation stagnation exceeds a preset threshold, at least one course optimization suggestion is automatically generated. The course optimization suggestion includes reducing the complexity of movements, extending the duration of a single training session, or adding assisted breathing training. The course optimization suggestions are pushed to the main interface, and the rehabilitation path is updated after the patient confirms.

[0008] Furthermore, the psychological assessment sub-interface includes an emotion assessment control and an emotion diary input area; In the psychological assessment sub-interface, an emotion assessment is performed. When an assessment result that matches a preset negative emotion pattern is detected, an emotion regulation mode is triggered, including the following steps: In response to the triggering operation of the emotion assessment control, a standardized depression or anxiety assessment scale is loaded and displayed, the patient's response to the selection of each item option is received, and an emotion score is generated. The patient's subjective description and classification of their emotional state on that day can be received through the text input area and emotion tag selector of the emotion diary input area. Based on the emotion score and the content of the emotion diary, when a combination that matches the preset negative emotion pattern is identified, the emotion regulation mode is triggered, the mindfulness training video playback window is automatically loaded, and emotion regulation suggestions are displayed, prompting the patient to confirm whether to start the intervention process.

[0009] Furthermore, the AR training sub-interface includes an AR guidance overlay area and gamified feedback controls; In the AR training sub-interface, food or utensils in the field of view are identified in real time, swallowing risk warnings and safe eating guidelines are displayed, and an interactive swallowing game is loaded. Feedback is provided based on the results of swallowing actions, and achievement badges and incentive information are updated, including the following steps: The device's camera captures images of food or tableware within the current field of view, and the recognition results are obtained. Based on the recognition results, a swallowing risk level label and corresponding safe eating operation instructions are generated in the AR guidance overlay area, and displayed as an augmented reality layer overlaid on the real-time image of the food or tableware. In response to the triggering operation of the gamified feedback control, an interactive training task synchronized with the swallowing action is loaded, the swallowing action execution signal is received, real-time visual or audio feedback is output according to the accuracy of the action, and achievement marks and incentive information are updated. If the swallowing integrity score is lower than the preset threshold in two consecutive interactive training tasks, and the patient's emotional label is identified as "frustration" or "anxiety", the difficulty of the next interactive training task will be automatically reduced.

[0010] Furthermore, the main interface also includes a rehabilitation progress visualization area, which includes: A line graph showing the changes over time based on the swallowing function scale score; The progress bar for achieving each rehabilitation goal is calculated by weighting the completion rate of the corresponding course and the accuracy of the movements. An achievement graph displays unlocked training levels and incentive badges; The rehabilitation progress visualization area allows patients to initiate course adjustment requests or goal revision requests by clicking on any chart element, thereby re-executing the rehabilitation path generation logic.

[0011] Furthermore, the method also includes a healthcare collaboration step: The patient information report interface is displayed on the medical staff's end, and the patient's swallowing function assessment report, rehabilitation path, training completion rate, emotion score and AR training feedback data are synchronized in real time. In response to the instructions from medical staff, suggestions for adjusting dietary plans, corrected training videos, or high-risk warning zones are pushed to the patient's main interface.

[0012] Furthermore, the method also includes a family collaboration step: After obtaining authorization from the patient, grant family members access to view training records, edit diet logs, and synchronize rehabilitation goal dashboards. It receives images of the eating environment uploaded by family members, identifies potential sources of risk, and automatically generates suggestions for environmental modifications.

[0013] On the other hand, this application provides an interactive system for rehabilitation training of dysphagia in stroke, including a patient terminal, a medical care terminal, a family member terminal and a cloud server terminal, for executing the aforementioned interactive method for rehabilitation training of dysphagia in stroke. The cloud server is used to enable data communication between the patient, the medical staff, and the family members.

[0014] The beneficial effects of this application are as follows: This application provides an interactive method for rehabilitation training of dysphagia in stroke patients. This method integrates four core modules: swallowing function assessment, personalized curriculum implementation, psychological state monitoring, and AR gamified training, constructing a closed-loop, multi-dimensional rehabilitation interactive system. Patients can complete a structured assessment of their swallowing ability and family support environment in one stop on the main interface. Based on this, the system generates an adjustable personalized rehabilitation path and dynamically collects execution feedback during training to optimize the curriculum content. Simultaneously, the emotion assessment mechanism can promptly identify negative psychological states and automatically trigger regulatory interventions, effectively alleviating the interference of anxiety or depression on rehabilitation. Combined with augmented reality technology, the system can also identify food or utensils in real-time in realistic eating scenarios, overlaying risk warnings and safety guidelines, and enhancing the fun and compliance of training through interactive games. Ultimately, it achieves a personalized, visual, emotional, and immersive integration of the rehabilitation process, significantly improving the rehabilitation effect and quality of life for stroke patients with dysphagia. This application also provides a corresponding system, the beneficial effects of which are similar to the method and will not be elaborated here.

[0015] Other features and advantages of this application will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the description, claims and drawings. Attached Figure Description

[0016] The accompanying drawings are provided to further understand the technical solutions of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the technical solutions of the present invention, and do not constitute a limitation on the technical solutions of the present invention.

[0017] Figure 1 This is a flowchart of the interactive method for rehabilitation training of dysphagia after stroke provided in this application; Figure 2 This is a schematic diagram of the main interface of the patient terminal for the rehabilitation training interaction for stroke swallowing disorders provided in this application; Figure 3 This is a schematic diagram of the swallowing function assessment sub-interface provided in this application; Figure 4 This is a schematic diagram of the course execution sub-interface provided in this application; Figure 5 This is a schematic diagram of the psychological assessment sub-interface provided in this application; Figure 6 This is a schematic diagram of the AR training sub-interface provided in this application; Figure 7 This is a structural diagram of the interactive rehabilitation training system for stroke swallowing disorders provided in this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0019] The present application will be further described below with reference to the accompanying drawings and specific embodiments. The described embodiments should not be considered as limitations on the present application, and all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of the present application.

[0020] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0022] Stroke is a leading cause of disability and death worldwide, with dysphagia being a common complication, particularly prevalent in acute-phase patients. Dysphagia not only severely impacts nutrient intake and fluid balance but also significantly increases the risks of aspiration, aspiration pneumonia, dehydration, and malnutrition, thereby prolonging hospital stays, increasing readmission rates, and reducing overall quality of life. Therefore, systematic and continuous rehabilitation interventions for dysphagia after stroke are crucial.

[0023] Currently, the clinical management of dysphagia mainly relies on face-to-face assessment and training by speech therapists or rehabilitation physicians. Commonly used tools include standardized swallowing function assessment methods such as the GUSS scale and the Kubota water swallowing test. Although these methods have a certain degree of reliability and validity, their implementation is highly dependent on professionals, and the assessment results are often static and single-point, making it difficult to dynamically track changes in the patient's actual swallowing ability in the home environment. In addition, traditional rehabilitation training is mainly passive, with patients mostly in a state of receiving instructions, lacking a sense of active participation and purpose, resulting in poor compliance and a high rate of training interruption, especially in the follow-up care stage after discharge.

[0024] In recent years, with the development of mobile health technology, some rehabilitation mini-programs or digital platforms have begun to offer functions such as online courses, check-in records, and video guidance. However, these existing digital solutions generally have several key shortcomings. First, the training content is mostly a general template, failing to fully consider individual patient differences, such as swallowing function level, psychological state, and family care conditions, and lacking a truly personalized pathway generation mechanism. Second, most existing systems focus only on the recovery of physiological functions, neglecting the profound impact of psychological factors on the rehabilitation process.

[0025] Empowerment theory is a health promotion concept centered on enhancing individual autonomy, sense of control, and self-efficacy. It emphasizes that in the healthcare process, patients should not merely be passive recipients of services, but rather active participants in decision-making and self-management. This theory advocates providing appropriate information, skills support, and resource links to help individuals identify their needs, set achievable goals, and master coping strategies. Through continuous feedback and practice, individuals enhance their control over their health behaviors. The empowerment process not only focuses on the restoration of physiological functions but also emphasizes the adjustment of psychological states and the construction of social support networks. Through doctor-patient collaboration, family involvement, and multidisciplinary collaboration, it creates a supportive environment that respects patients' wishes, stimulates intrinsic motivation, and promotes long-term behavioral change, ultimately achieving a fundamental shift from dependence on external intervention to autonomous health management.

[0026] To address the aforementioned issues, and based on the principles of empowerment theory, this application provides an interactive method and system for rehabilitation training of stroke-related dysphagia. It deeply integrates empowerment theory into the entire digital interactive process of stroke-related dysphagia rehabilitation, constructing a patient-centered, multi-dimensional collaborative intelligent rehabilitation system. The system links swallowing function assessment controls with family support questionnaires to achieve structured collection of physiological and social needs, and supports patients in independently confirming or adjusting rehabilitation goals, generating personalized training paths. The course execution module not only provides video-based training content but also dynamically optimizes the difficulty and content based on patient feedback, forming a closed-loop iteration. The psychological assessment module integrates standardized emotion scales and emotion diaries, automatically identifying negative emotion patterns and triggering mindfulness intervention. The AR gamified training module uses augmented reality technology to overlay risk warnings and safety guidelines in real eating scenarios, and enhances training immersion and compliance through interactive games and achievement incentives. Simultaneously, the system connects data channels between patients, medical staff, and family members, supporting multidisciplinary remote collaboration, authorized family participation, and intelligent environmental risk identification, truly achieving the integrated fusion of physiological rehabilitation, psychological empowerment, family collaboration, and professional support.

[0027] First, the interactive method for rehabilitation training of dysphagia after stroke provided in this application will be described in detail below with reference to the accompanying drawings.

[0028] Reference Figures 1 to 6The implementation process of the interactive method for rehabilitation training of dysphagia in stroke provided in this application includes, but is not limited to, the following steps.

[0029] Step S110: Display the main interface 100 of the patient terminal for stroke dysphagia rehabilitation training interaction.

[0030] Among them, reference Figure 2 The main interface 100 includes swallowing function assessment control 101, course execution control 102, psychological assessment control 103, and AR gamified training control 104.

[0031] In step S110, the patient-side main interface 100 for stroke swallowing disorder rehabilitation training is displayed. This interface integrates multiple core controls, including a swallowing function assessment control 101, a course execution control 102, a psychological assessment control 103, and an AR gamified training control 104. Through this integrated interface, patients can easily access different types of rehabilitation resources and services without switching between multiple applications or websites, greatly improving ease of use and efficiency. This one-stop design helps improve patient participation and adherence, making the rehabilitation process more systematic and easier to manage.

[0032] Step S120: In response to the triggering operation of the swallowing function assessment control 101, the swallowing function assessment sub-interface 200 is displayed, the patient's structured data in the swallowing function dimension and family support dimension are collected, a swallowing function assessment report is generated and a set of rehabilitation goals is determined, the patient's confirmation or adjustment instructions for the set of rehabilitation goals are received, and a rehabilitation path is generated.

[0033] The rehabilitation pathway includes a training course sequence and its difficulty parameters.

[0034] In step S120, in response to the operation of the swallowing function assessment control 101, a corresponding sub-interface is displayed and data is collected, covering both swallowing function and family support dimensions, thereby generating a detailed swallowing function assessment report and determining a personalized set of rehabilitation goals. This step is crucial because it not only helps the medical team fully understand the patient's current condition but also provides a basis for developing a scientific and reasonable rehabilitation plan. By involving the patient in the process of confirming or adjusting rehabilitation goals, their initiative can be enhanced, making the rehabilitation path more aligned with individual needs and maximizing rehabilitation effectiveness.

[0035] In step S130, in response to the triggering operation of the course execution control 102, the course execution sub-interface 300 is displayed, the training course corresponding to the rehabilitation path is loaded, the execution feedback and obstacle description submitted by the patient are obtained after the training is completed, and the training course is adjusted based on the execution feedback and obstacle description.

[0036] In step S130, in response to the triggering operation of the course execution control 102, training course content matching the patient's specific rehabilitation path is displayed. After the training is completed, patient feedback and descriptions of the obstacles encountered are collected. Based on this information, subsequent training plans are adjusted to ensure that the rehabilitation process always conforms to the patient's actual progress and needs. This step emphasizes the importance of personalization and dynamic adjustment. By obtaining real-time training feedback from the patient to optimize the training content, it effectively solves the problem that fixed patterns in traditional rehabilitation training cannot adapt to individual differences, enhancing the accuracy and flexibility of rehabilitation intervention.

[0037] In step S140, in response to the triggering operation of the psychological assessment control 103, the psychological assessment sub-interface 400 is displayed, and an emotion assessment is performed. When an assessment result that matches the preset negative emotion pattern is detected, the emotion regulation mode is triggered.

[0038] In step S140, an emotion assessment is initiated via psychological assessment control 103 to identify whether the patient is experiencing negative emotional states and to automatically activate the emotion regulation mechanism when necessary. This step focuses on the patient's mental health, recognizing the significant impact of emotional fluctuations on rehabilitation outcomes. Timely emotion assessment and intervention can help alleviate negative emotions such as anxiety and depression, improve the patient's mental state, and thereby enhance their enthusiasm and effectiveness in participating in rehabilitation training. This approach integrates mental health into the overall rehabilitation plan, embodying the concept of holistic mind-body treatment.

[0039] In step S150, in response to the triggering operation of the AR gamified training control 104, the AR training sub-interface 500 is displayed, which identifies food or tableware in the field of vision in real time, displays swallowing risk warnings and safe eating guidelines, loads swallowing interactive games, provides feedback based on the execution results of swallowing actions, and updates achievement marks and incentive information.

[0040] In step S150, in response to the operation of the AR gamified training control 104, the system enters the AR training sub-interface 500, uses augmented reality technology to identify food or utensils, provides safe eating guidance, and initiates an interactive game. Real-time feedback is provided based on the accuracy of swallowing movements, while achievement markers and incentive information are updated. This innovative approach combines rehabilitation training with everyday life scenarios, increasing the fun and immersion of the training. It not only helps improve patients' training interest and sustained participation but also allows them to apply learned skills in real life, strengthening the training effect and ultimately achieving better rehabilitation outcomes.

[0041] In some embodiments of this application, reference is made to Figure 2This invention provides a patient-side main interface 100 for rehabilitation training of dysphagia in stroke patients. This interface, based on the concept of empowerment, integrates multi-dimensional rehabilitation information and core functional modules, enabling patients to visualize and actively participate in their rehabilitation process. The central area of ​​the main interface 100 features a rehabilitation progress visualization area 105, including a line graph on the left showing the swallowing function score over time. The horizontal axis represents the date (e.g., October 3rd to November 15th), and the vertical axis represents the swallowing function score (from 45 points to over 95 points). The graph is marked with "Rapid Progress," "Steady Progress," and "Maintaining" stages, intuitively reflecting the functional recovery trend. The right side displays a progress bar group for the current goals, listing four rehabilitation goals: "Safe Drinking," "Tongue Muscle Training," "Laryngeal Elevation," and "Eating Tolerance." Each goal is accompanied by a weighted progress value (e.g., 95%, 85%, 93%, 98%), which is updated in real time. Below is a cumulative achievement graph, presented in a tree structure. The system displays unlocked training levels and incentive badges in a path format, such as "[Start Challenge] → [Level 1] → [Level 2] → [Badge A] → [Level 3] → [Level 4] → [Badge B]", forming a positive incentive loop. At the bottom of the interface, four core functional controls are arranged horizontally: "Swallowing Function Assessment," "Course Implementation," "Psychological Assessment," and "AR Gamified Training." Users can click on any control to enter the corresponding sub-interface for operation. All data is synchronized through a cloud server, allowing patients to independently confirm or adjust their rehabilitation goals after completing the assessment, thereby generating a personalized rehabilitation path. This fully embodies a patient-centered, dynamic feedback, and intelligently optimized digital rehabilitation interaction model.

[0042] In some embodiments of this application, reference is made to Figure 3 The swallowing function assessment sub-interface 200 includes a swallowing function scale control 201 and a family support questionnaire control 202. In step S120, in the swallowing function assessment sub-interface 200, structured data of the patient in the swallowing function dimension and family support dimension are collected, a swallowing function assessment report is generated, and a set of rehabilitation goals is determined. The patient's confirmation or adjustment instructions for the set of rehabilitation goals are received, and a rehabilitation path is generated, including the following steps.

[0043] In step S210, in response to the triggering operation of the swallowing function scale control 201, the swallowing function assessment scale is loaded and displayed, and the patient's score input for each swallowing ability item is received as swallowing function information.

[0044] In step S210, subjective self-assessment data of patients on swallowing physiological function are systematically collected, covering key dimensions such as chewing, laryngeal movement, cough reflex, and choking when drinking. This transforms the content that originally relied on on-site assessment by professionals into structured input that patients can complete independently, which not only improves the accessibility of assessment but also provides an objective basis for setting subsequent personalized rehabilitation goals.

[0045] In step S220, in response to the triggering operation of the family support questionnaire control 202, the structured questions of the home rehabilitation environment are loaded and displayed, and inputs about caregiver configuration and home training conditions are received as family support information.

[0046] In step S220, the focus is on the social support environment of the patient. By collecting key information through standardized questions, such as whether there is a dedicated person in the family to assist with training, whether the living space is convenient for practice, and whether the dietary preparation conditions are suitable, the family, an important rehabilitation resource, is included in the assessment system to ensure that the rehabilitation plan developed subsequently has practical feasibility and a basis for implementation.

[0047] Step S230: Based on swallowing function information and family support information, generate a swallowing function assessment report and construct a set of rehabilitation goals accordingly.

[0048] In step S230, based on swallowing function information and family support information, and by integrating data on both physiological ability and social support, a comprehensive assessment report is generated that fully reflects the patient's current rehabilitation starting point. On this basis, a set of clear and feasible rehabilitation goals is automatically generated. These goals take into account both the patient's degree of functional impairment and the family's caregiving ability, laying a scientific foundation for subsequent path planning.

[0049] Optionally, by numerically encoding the structured data from the swallowing function scale and classifying and encoding the unstructured data from the family support questionnaire, a weighted fusion algorithm or machine learning model (such as logistic regression, random forest, or gradient boosting tree) is used to combine physiological function scores with social support scores to generate a comprehensive assessment report. This is applied to scenarios such as patients' first use of the system or regular assessments to collect data on swallowing physiological abilities and family care conditions. This technology overcomes the limitations of single physiological assessments by incorporating family environmental factors, making the assessment results more closely aligned with the patient's actual situation, providing a scientific basis for setting subsequent rehabilitation goals, and improving the feasibility of those goals.

[0050] Furthermore, in the process of generating swallowing function assessment reports and constructing rehabilitation goal sets, rule engines and knowledge graph technologies can be combined. Clinical guidelines (such as the Chinese Expert Consensus on Swallowing Disorder Rehabilitation) can be transformed into a structured rule base and matched with individual patient data to automatically generate an initial goal set that conforms to evidence-based medicine. At the same time, natural language generation technologies (such as the BERT language processing model or LLM large language model) can be used to translate the assessment results into easily understandable patient language, improving information accessibility.

[0051] Step S240: Display the set of rehabilitation goals in the swallowing function assessment sub-interface 200, receive the patient's confirmation operation or adjustment instruction for the rehabilitation goals, and generate a rehabilitation path based on the confirmed set of rehabilitation goals.

[0052] Step S240 emphasizes the patient's principal role and right to participate in decision-making, allowing the patient to review, confirm, or modify the goals that were originally set unilaterally by medical staff, making them a true co-creator of the rehabilitation process; once the goals are determined, the system generates a rehabilitation path that includes a specific training course sequence and its difficulty parameters, achieving a seamless connection from assessment to intervention, reflecting the core principle of empowerment.

[0053] In some embodiments of this application, reference is made to Figure 3 A swallowing function assessment sub-interface 200 is provided to collect structured data on the swallowing ability and family support environment of stroke patients during rehabilitation and to generate a preliminary set of rehabilitation goals. The interface contains two core assessment modules in the middle: the first module is the "Swallowing Function Scale," which prompts the user to "score according to the actual situation of the last 3 days" and is marked "[Completed]," indicating that the scale has been completed and is used to quantify the patient's swallowing function status; the second module is the "Family Support Questionnaire," which prompts the user to "answer questions about care and home conditions," and is also marked "[Completed]," used to collect information on the care support and home training conditions that the patient receives in the home environment; the bottom of the interface displays a preliminary list of rehabilitation goals 203 automatically generated by the system based on the above two assessment results, including "Safe Drinking Techniques (2 weeks)," "Improved Tongue Muscle Control (3 weeks)," and "Enhanced Oral Feeding Tolerance (4 weeks)," with a "[Click to View]" link after each goal for the user to view detailed instructions; two operation buttons are set at the bottom: "Confirm Goals" and "Adjust Goals," allowing the patient to make final confirmations or personalized modifications to the goals generated by the system, thereby realizing a closed-loop interactive process from assessment to goal setting, fully reflecting the patient's active participation and decision-making autonomy in the development of the rehabilitation plan.

[0054] In some embodiments of this application, reference is made to Figure 4 The course execution sub-interface 300 includes a course content display area 301, a training check-in control 302, and a feedback input control 303. In step S130, the training course corresponding to the rehabilitation path is loaded in the course execution sub-interface 300. After the training is completed, the execution feedback and obstacle description submitted by the patient are obtained, and the training course is adjusted based on the execution feedback and obstacle description, including the following steps.

[0055] Step S310: In the course content display area 301, load and display the training course corresponding to the rehabilitation path, presenting the key points of swallowing training movements in the form of video or text.

[0056] In step S310, the abstract training plan in the personalized rehabilitation pathway is transformed into specific, intuitive, and easy-to-understand guidance content. Through standardized video demonstrations or graphic explanations, it is ensured that patients can accurately grasp the technical points and execution standards of each swallowing training movement, thereby improving the correctness and effectiveness of the training and avoiding the reduction of rehabilitation effects or even the risk caused by movement deviations.

[0057] In step S320, in response to the trigger operation of the training check-in control 302, the completion status of the current course is recorded, and subsequent operations are locked until feedback is submitted.

[0058] In step S320, by establishing a mandatory association mechanism between training completion and feedback submission, the system prevents patients from simply completing the check-in and neglecting to reflect on and report on the training process. By temporarily locking subsequent functions, the system guides patients to immediately enter the feedback stage after completing the training, ensuring the timeliness and completeness of feedback data, providing a reliable basis for subsequent course optimization, and also strengthening patients' focus on training quality rather than just on the formal completion.

[0059] In step S330, in response to the triggering operation of the feedback input control 303, a training feedback window is displayed to receive selective input from the patient regarding the training execution status. The selective input includes the degree of action completion, the type of subjective difficulty, and a description of the body's reaction. The feedback data is then linked to the current course record, and the content of the training course is adjusted accordingly.

[0060] In step S330, by constructing a dynamic feedback channel centered on patient experience, the system can obtain first-hand information about the actual performance of training. By collecting multi-dimensional data such as the degree of action completion, specific difficulties encountered, and physical reactions in a structured manner, the system can accurately identify obstacles in training and make targeted adjustments to subsequent course content based on this, achieving truly personalized iteration and adaptive rehabilitation.

[0061] In some embodiments of this application, reference is made to Figure 4This system provides a course execution sub-interface 300 to guide stroke patients in completing personalized swallowing rehabilitation training. The main functional areas are marked with dotted lines in the central area of ​​the interface. The current course information is displayed at the top, including the course name "Tongue Muscle Protrusion Training - Lesson 3," the goal description "Improve tongue muscle control," and the difficulty level indicator "Beginner." A video playback window in the center displays demonstration videos of standard training movements to help patients accurately grasp the key points. The right side lists the instructions for the movements: "Tongue tip against the upper gum, slowly protrude forward, hold for 3 seconds, repeat 10 times," assisting users in understanding and imitating the correct movements. At the bottom of the interface, two core interactive controls are arranged horizontally: a training check-in control 302 on the left, used to record the completion status of this training session; and a feedback input control 303 on the right, used to guide patients to submit their execution status and descriptions of obstacles during the training process. This allows the system to dynamically optimize subsequent training content based on feedback data, achieving closed-loop management from learning to execution to feedback, ensuring the standardization, traceability, and personalized adjustment capabilities of the training.

[0062] In some embodiments of this application, during step S230, which generates a swallowing function assessment report and constructs a rehabilitation goal set, rule engines and knowledge graph technologies can be combined. Clinical guidelines (such as the Chinese Expert Consensus on Swallowing Disorder Rehabilitation) are transformed into a structured rule base and matched with individual patient data to automatically generate an initial goal set that conforms to evidence-based medicine. Simultaneously, natural language generation technology is used to translate the assessment results into easily understandable patient language, improving information accessibility.

[0063] In some embodiments of this application, step S330 involves associating feedback data with the current course record and adjusting the content of the training course accordingly, specifically including the following steps.

[0064] Step S410: Using a pre-set rehabilitation progress prediction model, the patient's historical training completion rate, the frequency of obstacle type distribution, and the trend of emotion score change are used as input features to output the patient's rehabilitation stagnation probability.

[0065] In step S410, the patient's rehabilitation process is quantitatively assessed through a data-driven approach. Instead of relying on subjective judgment, the assessment integrates multi-dimensional dynamic information such as the patient's past training behavior, common difficulty patterns, and changes in psychological state to construct an objective indicator that reflects the potential for rehabilitation progress. This allows for the early identification of potential bottlenecks or insufficient motivation, providing a scientific basis for timely intervention.

[0066] Optionally, a time-series prediction model, such as a Long Short-Term Memory Network (LSTM) or a Transformer-based time series model, can be used as a rehabilitation progress prediction model to effectively capture the dynamic evolution of rehabilitation behavior. Alternatively, a multi-task learning framework can be integrated to jointly predict the speed of functional recovery, dropout risk, and emotional fluctuations, thereby achieving a more comprehensive progress assessment.

[0067] Step S420: When the probability of rehabilitation stagnation exceeds a preset threshold, at least one course optimization suggestion is automatically generated. The course optimization suggestion includes reducing the complexity of movements, extending the duration of a single training session, or adding assisted breathing training.

[0068] In step S420, the risk identification of rehabilitation stagnation is transformed into specific and actionable intervention strategies, enabling the system to proactively optimize. When the model determines that the patient may be stuck in a rehabilitation plateau or face difficulties in execution, the system no longer maintains the original training plan, but intelligently recommends targeted adjustment measures, such as simplifying movements to reduce frustration, increasing training time to strengthen consolidation, or introducing auxiliary training to improve overall coordination, thereby breaking the stagnation and maintaining rehabilitation momentum.

[0069] Step S430: Push the course optimization suggestions to the main interface 100, and wait for the patient's confirmation before updating the rehabilitation path.

[0070] In step S430, the patient's primary role in rehabilitation adjustments is emphasized. Although the system can automatically generate optimized plans, it does not enforce their implementation. Instead, the suggestions are clearly presented on the main interface 100, allowing the patient to review and decide whether to adopt them. This mechanism ensures both the timeliness and professionalism of the intervention and respects the patient's wishes and sense of control. It ensures that every adjustment to the rehabilitation pathway is based on doctor-patient collaboration or the patient's self-approval, thereby enhancing the patient's acceptance of the new plan and willingness to implement it, and achieving dynamic co-governance under the concept of empowerment.

[0071] In some embodiments of this application, reference is made to Figure 5 The psychological assessment sub-interface 400 includes an emotion assessment control 401 and an emotion diary input area 402. In step S140, an emotion assessment is performed in the psychological assessment sub-interface 400. When an assessment result that matches a preset negative emotion pattern is detected, an emotion regulation mode is triggered, including the following steps.

[0072] In step S510, in response to the triggering operation of the emotion assessment control 401, a standardized depression or anxiety assessment scale is loaded and displayed, the patient's check-in response for each item option is received, and an emotion score is generated.

[0073] In step S510, the patient's psychological state is assessed in a structured and quantitative manner using clinically validated standardized tools, transforming the originally subjective emotional experience into measurable and traceable numerical indicators. In the process of completing the scale independently, the patient not only achieves self-awareness but also provides the system with objective evidence to determine whether there is a significant tendency towards depression or anxiety, laying the foundation for whether subsequent intervention is needed.

[0074] In step S520, the patient's subjective description and classification of their emotional state on that day are received through the text input area and emotion tag selector of the emotion diary input area 402.

[0075] In step S520, the patient's active input is used to supplement the deficiencies of the structured scale and capture the patient's immediate, subtle and personalized emotional fluctuations in daily life. By combining free text recording and preset emotion tags, the flexibility of the patient's expression is preserved, and the system can classify and analyze the emotion types, thereby forming a more comprehensive and dynamic emotional profile and enhancing the depth and realism of psychological state recognition.

[0076] Step S530: Based on the emotion score and the content of the emotion diary, when a combination that matches the preset negative emotion pattern is identified, the emotion regulation mode is triggered, the mindfulness training video playback window is automatically loaded, and emotion regulation suggestions are displayed, prompting the patient to confirm whether to start the intervention process.

[0077] In step S530, when the system comprehensively judges that the patient is in a high-risk emotional state, it no longer stops at the level of data recording, but actively pushes emotion regulation resources with clinical evidence to realize the automation and contextualized response of psychological intervention. By presenting mindfulness training videos and specific suggestions in real time, and giving patients the right to choose whether to accept intervention, it not only reflects the importance attached to mental health, but also practices the principle of empowering patients to manage their emotions autonomously.

[0078] Optionally, a standardized scale is used to generate an emotion score, and the emotion diary text is classified using the BERT text sentiment analysis model. The score and classification results are then combined to identify negative emotion patterns. Mindfulness training videos are automatically pushed when regulation is triggered. This technology can be applied to scenarios where patients are conducting psychological assessments or recording their daily emotions. This technology accurately captures patients' emotional fluctuations, automatically provides psychological intervention resources, enhances patients' psychological resilience, and reduces the impact of negative emotions on rehabilitation.

[0079] In some embodiments of this application, reference is made to Figure 5This system provides a psychological assessment sub-interface 400 for systematically collecting the subjective emotional state of stroke patients and implementing intelligent intervention responses. The interface contains two core modules: the first module is "Emotional Assessment," prompting the user to "answer based on your feelings over the past week," and marking it "[Completed]," indicating that a standardized emotional scale has been completed for quantitative assessment of the patient's mental health; the second module is "Today's Mood Record," including a text input area and an emotion tag selector, displaying example content such as "Today's Mood: Bad weather, difficult to eat," and marking it with the emotion tag "Sad," supporting patients to express their daily emotional experiences in a structured manner; a dynamic prompt box 403 is set at the bottom of the interface. When the system identifies a preset negative emotion pattern, it automatically pops up a prompt message: "We noticed that you have been feeling somewhat down recently. Would you like to try a mindfulness relaxation training?" Below this, two operation options are provided: "Try Now" and "Remind Me Later," allowing the patient to decide whether to initiate the emotion regulation intervention process, achieving closed-loop management from emotion monitoring to proactive intervention, fully embodying a comprehensive rehabilitation concept centered on the patient and taking into account both physiological and psychological needs.

[0080] In some embodiments of this application, reference is made to Figure 6 The AR training sub-interface 500 includes an AR guidance overlay area 501 and a gamified feedback control 502. In step S150, in the AR training sub-interface 500, food or tableware in the field of vision is identified in real time, swallowing risk warnings and safe eating guidelines are displayed, and a swallowing interactive game is loaded. Feedback is provided based on the execution results of the swallowing action, and achievement icons and incentive information are updated, including the following steps.

[0081] Step S610: Use the device's camera to capture images of food or tableware in the current field of view and obtain recognition results.

[0082] In step S610, the real eating environment is used as the input scenario for rehabilitation training. The device's camera acquires real-time visual information of the food or tableware in front of the patient, and image recognition technology is used to classify and analyze these objects. This process is the foundation for all subsequent augmented reality interactions and risk warnings, enabling the system to understand the specific eating objects that the patient is currently facing, thus providing a prerequisite for personalized and contextualized swallowing guidance.

[0083] Optionally, images are captured using the device's camera, and the YOLOv8 object detection model is used to detect the categories and locations of food and utensils. This information is then combined with a pre-defined risk level library to generate risk labels and safe eating guidelines. This is applied to scenarios where patients use AR training controls for eating practice. This technology achieves real-time and accurate scene recognition, provides timely guidance, reduces the risk of aspiration, and enhances the realism and practicality of the training scenario.

[0084] Step S620: Based on the recognition results, a swallowing risk level label and corresponding safe eating operation instructions are generated in the AR guidance overlay area 501, and displayed as an augmented reality layer overlaid on the real-time image of the food or tableware.

[0085] In step S620, abstract knowledge of swallowing safety is transformed into intuitive, immediate, and visually integrated guidance embedded in real-world scenarios. Based on the characteristics of the identified food or the type of tableware, the system automatically determines the swallowing difficulty and potential risks, and directly overlays operation prompts in the form of text, icons, or animations in the patient's field of vision, such as suggesting small bites, adjusting posture, or avoiding specific textures. This provides precise and imperceptible assistance during actual eating, improving the practical ability to eat safely.

[0086] In step S630, in response to the triggering operation of the gamified feedback control 502, an interactive training task synchronized with the swallowing action is loaded, the swallowing action execution signal is received, real-time visual or audio feedback is output according to the accuracy of the action, and achievement marks and incentive information are updated.

[0087] In step S630, rehabilitation training is transformed into an interactive and fun game experience. The system simultaneously captures the patient's behavior when performing swallowing actions and judges whether the actions are standardized in real time according to preset standards. It then provides positive or corrective feedback through visual effects, sound effects, or prompts. At the same time, it records training results and updates incentive elements such as badges, points, or level status, thereby enhancing the patient's sense of participation, sense of accomplishment, and motivation for continuous training.

[0088] In step S640, when the swallowing integrity score is lower than the preset threshold in two consecutive interactive training tasks, and the patient's emotional label is identified as "frustration" or "anxiety", the difficulty of the next interactive training task is automatically reduced.

[0089] Step S640 demonstrates the system's collaborative perception and intelligent response to the patient's physiological performance and psychological state. When the system detects that the training performance is consistently poor and there are negative emotional markers, it actively reduces the complexity or requirements of subsequent tasks to reduce cognitive load and psychological pressure, and prevents repeated failures from exacerbating feelings of frustration. This maintains the continuity of training while protecting the patient's sense of self-efficacy and realizes an emotionally sensitive adaptive regulation mechanism.

[0090] In some embodiments of this application, reference is made to Figure 6An AR training sub-interface 500 is provided to integrate augmented reality technology and gamified interaction in real eating scenarios, thereby enhancing the immersion and compliance of stroke patients in swallowing rehabilitation training. The central area of ​​the interface is marked with a dotted frame to indicate the core interactive area, with a message above stating "Camera is on, microphone is on," indicating that the device's sensors are working. The central area displays a real-time camera feed overlaid with AR game elements, showcasing the food or utensils in front of the patient and superimposing dynamic visual information. The right side displays the current training task, such as "Swallowing Challenge: Soup Level," with the objective of "Complete 3 safe swallows," and provides feedback such as "Successfully completed two, great job!" Swallowing tips are also provided: "Rice is more difficult to swallow, requiring small bites and swallowing with your head down; soup is easier to swallow and can be consumed normally," providing contextualized safety guidance. At the bottom of the interface is a gamified feedback control 502, used to receive real-time evaluations of the patient's swallowing actions. It supports outputting positive incentives or corrective suggestions through visual or audio means, combined with task progress and achievement mechanisms, enhancing the fun and motivation for continued participation, achieving a closed-loop intelligent rehabilitation training from real-world environmental perception to virtual feedback guidance.

[0091] In some embodiments of this application, reference is made to Figure 2 The main interface 100 also includes a rehabilitation progress visualization area 105, which includes the following content.

[0092] (1) The line graph generated based on the swallowing function scale score over time presents the recovery process of the patient's swallowing function in an intuitive and continuous visual form. By regularly recording and updating the scale score data, the system can draw a dynamic curve that reflects the recovery trend, so that the patient can clearly see the time nodes and amplitude of their own functional improvement or fluctuation, thereby enhancing their perception and confidence in the recovery process, and also providing an objective basis for self-reflection and goal adjustment.

[0093] (2) Progress bars for achieving various rehabilitation goals, where the progress value is calculated by weighting the completion rate of the corresponding course and the accuracy of the movement, going beyond simple task completion statistics and comprehensively considering the quantity and quality of training; the progress bar not only reflects whether the patient has completed the designated course, but also incorporates standardized indicators of movement execution, making the assessment of goal progress more comprehensive and realistic; this quantitative and multi-dimensional feedback mechanism helps patients understand that rehabilitation effectiveness depends not only on the frequency of participation, but also on the correctness of training, thereby guiding them to pay attention to the details of movement and improve the effectiveness of training.

[0094] (3) Achievement cumulative map, used to display unlocked training levels and incentive badges. The structured and visualized growth map enhances the patient's sense of accomplishment and long-term motivation to participate; it not only records the patient's training trajectory along the way, but also uses badges and other forms to reward key milestones, forming a positive incentive loop; this gamified narrative makes the rehabilitation process more meaningful and exploratory, helping patients to recall their existing achievements when facing difficulties and maintain their intrinsic motivation.

[0095] In some embodiments of this application, the rehabilitation progress visualization area 105 allows patients to initiate course adjustment requests or goal revision requests by clicking on any chart element, thereby re-executing the rehabilitation path generation logic. This empowers patients with proactive control over the rehabilitation process: whether questioning slow functional improvement at a certain stage or feeling uncomfortable with the current goal setting, patients can immediately express their adjustment requests through direct interaction with chart elements; the system then initiates a path reconstruction process, incorporating patient feedback into the decision-making loop, truly achieving user-centric, dynamically responsive, and empowering rehabilitation management.

[0096] In some embodiments of this application, the method further includes the following medical care collaboration steps.

[0097] Step S710: Display the patient information report interface on the medical staff terminal, and synchronize the patient's swallowing function assessment report, rehabilitation path, training completion rate, emotion score and AR training feedback data in real time.

[0098] In step S710, a comprehensive, dynamic, and centralized panoramic view of patient rehabilitation is provided to medical staff, enabling them to instantly grasp the multidimensional performance of patients in terms of physiological function, training compliance, psychological state, and actual eating scenarios without relying on patients' verbal reports or scattered records. This real-time data synchronization mechanism significantly improves the visibility and control of medical staff over the home rehabilitation process, laying the information foundation for remote monitoring, risk identification, and precise intervention.

[0099] In step S720, in response to the operation instructions of medical staff, suggestions for adjusting the diet plan, correcting training videos, or marking high-risk warning zones are pushed to the main interface 100 of the patient's terminal.

[0100] In step S720, an efficient and targeted channel for professional guidance is established to deliver information to patients, enabling healthcare professionals to provide personalized interventions based on the latest data. Whether it is adjusting the texture of food to reduce the risk of aspiration, correcting incorrect actions through demonstration videos, or marking specific time periods as high-risk periods to enhance vigilance, these professional opinions can reach patients directly in a structured and visual manner, ensuring that medical decisions are quickly translated into actionable actions for patients, thereby enhancing the effectiveness and safety of continuing care.

[0101] In some embodiments of this application, the method further includes the following family collaboration steps.

[0102] Step S810: After obtaining authorization from the patient, grant the family member access to view training records, edit diet logs, and synchronize rehabilitation goal dashboards.

[0103] In step S810, by constructing a family collaborative support mechanism based on the patient's wishes, and by sharing key rehabilitation information in a controlled manner, family members can substantially participate in the patient's daily care. Family members can understand the progress of training to provide supervision or encouragement, edit food logs to help record eating content and reactions, and simultaneously view the rehabilitation goal board to maintain consistency with the patient's goals and form a consensus. This permission design respects the patient's right to privacy and autonomy, activates the function of the family as a rehabilitation support unit, and strengthens the continuity and collaboration of home rehabilitation.

[0104] Step S820: Receive images of the eating environment uploaded by family members, identify potential risk sources, and automatically generate environmental modification suggestions.

[0105] In step S820, the home physical environment is incorporated into the rehabilitation risk management system. Using real-life photos taken by family members, the system uses image recognition technology to automatically detect risk factors that may increase aspiration or difficulty eating, such as sharp utensils, high-viscosity food containers, and poorly lit areas. Based on this, the system generates specific and actionable environmental optimization suggestions, such as replacing non-slip placemats, adjusting the table position, or using cups of a specific shape. This process transforms abstract safety guidelines into personalized prompts for real-life scenarios, effectively improving the scientific nature and preventative capabilities of home care.

[0106] Optionally, risk sources can be identified from the images of the eating environment uploaded by family members. Instance segmentation models (such as Mask R-CNN) can be used to accurately locate the edges of tableware, food spill areas, or dark areas. Combined with scene understanding models (SceneGraph Generation), potential dangerous relationships between objects in the environment (such as "a sharp fork near the corner of the mouth") can be inferred to generate more context-aware modification suggestions.

[0107] In some embodiments of this application, the system embeds technical features that proactively guide family members to provide support within the rehabilitation interaction process, thereby strengthening the synergistic role of family care in the rehabilitation of stroke-related dysphagia. When patients continuously report difficulties, low completion rates of actions, or negative emotional states such as "frustration" or "anxiety" during course execution or AR training, the system automatically pushes structured assistance prompts to authorized family members, including specific operational suggestions such as "It is recommended to accompany the patient to complete the next training session," "You can assist in adjusting the eating posture," or "Please confirm whether the current diet complies with safety guidelines." In addition, the system sets up a "Family Collaboration Tasks" module in the rehabilitation goal dashboard, such as "Assist in completing 3 AR safe eating exercises this week" or "Review the rehabilitation progress chart together," and promotes continuous family participation through message reminders and completion check-in mechanisms. When environmental images uploaded by family members are identified as high-risk factors, in addition to generating modification suggestions, the system also simultaneously sends positive communication messages such as "Your family member suggests changing to a non-slip placemat" to the patient's end, promoting positive interaction among family members. These technical features transform family members from passive observers to active collaborators, effectively enhancing the support strength and emotional connection of home rehabilitation.

[0108] In some embodiments of this application, the system fully incorporates age-friendly design principles into the main patient interface 100 and all sub-interfaces to improve the user experience and operational feasibility for elderly stroke patients. Specifically, this includes: using a high-contrast color scheme and large fonts to display control labels and operation prompts to ensure visual recognizability; all interactive buttons are at least 44×44 pixels in size and have sufficient spacing to avoid accidental touches; voice guidance is integrated throughout the swallowing function assessment, course execution, and AR training process, supporting one-click playback of the current step's instructions; voice prompts for questions and options are enabled in the swallowing function scale and emotion assessment interfaces, allowing patients to confirm via voice or complete input with simple clicks; the AR training sub-interface 500 simplifies gesture operations, requiring only gaze or a single click to initiate recognition and games; the rehabilitation progress visualization area 105 uses a dual approach of icons and text, with line graphs and progress bars supplemented by voice interpretation of key changes; simultaneously, the system supports a remote assistance mode for family members, who, upon authorization, can complete complex settings on behalf of the patient, while retaining the patient's ultimate control over core decisions such as goal confirmation and path adjustment. The aforementioned age-friendly guidance technology features effectively lower the digital usage threshold for elderly users and ensure their ability to participate autonomously in the empowerment and rehabilitation process.

[0109] Secondly, refer to Figure 7This application provides an interactive rehabilitation training system for stroke patients with dysphagia, comprising a patient terminal, a medical staff terminal, a family member terminal, and a cloud server terminal, for executing the aforementioned interactive rehabilitation training method for stroke patients with dysphagia. The system architecture aims to construct a multi-party, clearly defined, and functionally collaborative digital rehabilitation ecosystem, integrating patients, medical staff, and families into a unified technical platform, enabling the structured realization of their respective responsibilities and interactions during the rehabilitation process. The patient terminal focuses on self-training and self-management, the medical staff terminal emphasizes professional monitoring and remote intervention, and the family member terminal undertakes care support and environmental coordination. Through system division of labor and collaboration, the three parties jointly support the implementation of an empowerment-based rehabilitation model.

[0110] The cloud server is used to enable data communication between patients, medical staff, and family members. As the data hub and communication bridge of the entire system, the cloud server ensures real-time synchronization, secure transmission, and consistent storage of information across multiple devices. All assessment data, training records, emotional feedback, dietary logs, risk warnings, and interactive commands are aggregated, processed, and distributed through the cloud server. This breaks down the limitations of information silos in traditional rehabilitation, allowing relevant parties to promptly and accurately monitor the patient's rehabilitation status, whether in the hospital or at home. This provides a solid technical foundation for dynamically adjusting intervention strategies, achieving closed-loop management, and efficient collaboration.

[0111] In summary, the interactive method and system for rehabilitation training of dysphagia after stroke provided in this application have the following technical effects.

[0112] This application achieves one-stop integration of the rehabilitation process by constructing a patient-side main interface that integrates swallowing function assessment, personalized curriculum implementation, psychological state monitoring, and AR gamified training, significantly improving patient convenience and participation compliance. Based on structured data from both swallowing function and family support dimensions, it generates a negotiable set of rehabilitation goals and dynamically generates personalized rehabilitation paths containing course sequences and difficulty parameters, ensuring that the intervention plan is both clinically scientific and practically feasible. During training, a mandatory check-in and feedback mechanism is implemented, and a rehabilitation progress prediction model intelligently identifies the risk of stagnation, automatically generating and pushing optimization suggestions, achieving precise iteration of training content while respecting the patient's decision-making rights. Standardized emotion assessments and emotion diaries are integrated for analysis, automatically triggering mindfulness intervention and other emotion regulation modes, effectively alleviating the interference of negative psychology on rehabilitation and embodying the concept of mind-body synergy. Augmented reality technology is used to overlay risk warnings and safety guidelines in real eating scenarios, and through interactive games with synchronized movements and achievement incentive mechanisms, the tedious training is transformed into an immersive, contextualized experience, strengthening skill transfer and long-term motivation.

[0113] Furthermore, this application introduces a visualization area for rehabilitation progress, visually presenting the functional recovery trajectory with line graphs, weighted progress bars, and tree-like achievement maps. It also allows patients to click on charts to initiate path adjustment requests, truly empowering them to take control of their rehabilitation. Simultaneously, it establishes a collaborative channel between medical staff and family members, enabling real-time synchronization of assessment reports, training data, and emotional states. This supports professional remote intervention and home care collaboration, and uses image recognition to assist in environmental risk modification, building a continuous rehabilitation ecosystem. The overall system is based on empowerment theory, deeply integrating physiological, psychological, and family-related multidimensional needs. It promotes a fundamental shift in stroke swallowing disorder rehabilitation from passive execution to proactive co-treatment, from isolated intervention to multi-party collaboration, and from static planning to dynamic optimization, significantly improving rehabilitation efficiency, safety, and patients' quality of life.

[0114] It should be noted that in all specific embodiments of this application, all data processing activities related to user identity or personal characteristics, such as user information, user behavior data, historical data, and location information, will be conducted in accordance with the principles of legality, legitimacy, and necessity. All data collection, use, storage, and processing will be subject to compliance with applicable national and regional laws, regulations, and industry standards, and informed consent from users will be obtained in a clear and explicit manner before processing. For the processing of sensitive personal information, separate consent from users will be obtained through prominent means such as pop-up prompts and independent confirmation pages. If any processing conflicts with laws and regulations, the laws and regulations will prevail, and necessary data processing will only be carried out within the scope permitted by laws and regulations, ensuring that all data-based applications, analyses, and technical implementations are conducted within the scope permitted by laws and regulations.

[0115] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this application are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and sub-operations described as part of a larger operation are executed independently.

[0116] Furthermore, although this application is described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding this application. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of ordinary skill of an engineer. Therefore, those skilled in the art can implement the application set forth in the claims using ordinary skill. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of this application, which is determined by the full scope of the appended claims and their equivalents.

[0117] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several programs to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0118] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequential list of executable programs for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, a program execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can retrieve and execute a program from or in conjunction with such a program execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can mean any means that can contain, store, communicate, propagate, or transmit a program for use by or in conjunction with a program execution system, apparatus, or device.

[0119] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Additionally, computer-readable media can even be paper or other suitable media on which programs can be printed, for example, by optically scanning the paper or other media, then editing, interpreting, or, if necessary, processing it in a suitable manner to obtain the program electronically, and then storing it in computer memory.

[0120] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable program execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0121] In the foregoing description of this specification, the reference to terms such as "one embodiment / implementation," "another embodiment / implementation," or "certain embodiments / implementations," etc., indicates that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in an embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0122] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

[0123] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of the present invention.

Claims

1. An interactive method for rehabilitation training of dysphagia after stroke, characterized in that, Includes the following steps: The main interface of the patient terminal for stroke dysphagia rehabilitation training is displayed. The main interface includes swallowing function assessment controls, course execution controls, psychological assessment controls, and AR gamified training controls. In response to the triggering operation of the swallowing function assessment control, a swallowing function assessment sub-interface is displayed, structured data of the patient in the swallowing function dimension and family support dimension are collected, a swallowing function assessment report is generated and a set of rehabilitation goals is determined, the patient's confirmation or adjustment instructions for the set of rehabilitation goals are received, and a rehabilitation path is generated; the rehabilitation path includes a training course sequence and its difficulty parameters. In response to the triggering operation of the course execution control, the course execution sub-interface is displayed, the training course corresponding to the rehabilitation path is loaded, the execution feedback and obstacle description submitted by the patient are obtained after the training is completed, and the training course is adjusted based on the execution feedback and obstacle description. In response to the triggering operation of the psychological assessment control, the psychological assessment sub-interface is displayed, and an emotion assessment is performed. When an assessment result that matches a preset negative emotion pattern is detected, the emotion regulation mode is triggered. In response to the triggering operation of the AR gamified training control, an AR training sub-interface is displayed, which identifies food or tableware in the field of vision in real time, displays swallowing risk warnings and safe eating guidelines, loads a swallowing interactive game, provides feedback based on the execution result of the swallowing action, and updates achievement marks and incentive information.

2. The interactive method for rehabilitation training of dysphagia after stroke according to claim 1, characterized in that, The swallowing function assessment sub-interface includes swallowing function scale controls and family support questionnaire controls; In the swallowing function assessment sub-interface, structured data of the patient in the swallowing function dimension and family support dimension are collected, a swallowing function assessment report is generated, and a set of rehabilitation goals is determined. The system receives confirmation or adjustment instructions from the patient regarding the set of rehabilitation goals and generates a rehabilitation path, including the following steps: In response to a trigger operation on the swallowing function scale control, the swallowing function assessment scale is loaded and displayed, and the patient's score input for each swallowing ability item is received as swallowing function information. In response to the triggering operation of the family support questionnaire control, the structured questions of the home rehabilitation environment are loaded and displayed, and inputs about caregiver configuration and home training conditions are received as family support information; Based on the swallowing function information and the family support information, a swallowing function assessment report is generated, and a set of rehabilitation goals is constructed accordingly. The set of rehabilitation goals is displayed in the swallowing function assessment sub-interface, which receives the patient's confirmation or adjustment instructions for the rehabilitation goals and generates the rehabilitation path based on the confirmed set of rehabilitation goals.

3. The interactive method for rehabilitation training of dysphagia after stroke according to claim 1, characterized in that, The course execution sub-interface includes a course content display area, a training check-in control, and a feedback input control; In the course execution sub-interface, the training course corresponding to the rehabilitation path is loaded. After the training is completed, the execution feedback and obstacle description submitted by the patient are obtained, and the training course is adjusted based on the execution feedback and obstacle description, including the following steps: In the course content display area, training courses corresponding to the rehabilitation path are loaded and displayed, presenting the key points of swallowing training movements in the form of videos or text. In response to the triggering operation of the training check-in control, the completion status of the current course is recorded, and subsequent operations are locked until feedback is submitted; In response to the triggering operation of the feedback input control, a training feedback window is displayed to receive selective input from the patient regarding the training execution status. The selective input includes the degree of action completion, the type of subjective difficulty, and a description of the body's reaction. The feedback data is then associated with the current course record, and the content of the training course is adjusted accordingly.

4. The interactive method for rehabilitation training of dysphagia after stroke according to claim 3, characterized in that, The step of associating the feedback data with the current course record and adjusting the content of the training course accordingly specifically includes: Using a pre-defined rehabilitation progress prediction model, the model takes the patient's historical training completion rate, the frequency of the distribution of obstacle types, and the trend of changes in emotional scores as input features, and outputs the patient's rehabilitation stagnation probability. When the probability of rehabilitation stagnation exceeds a preset threshold, at least one course optimization suggestion is automatically generated. The course optimization suggestion includes reducing the complexity of movements, extending the duration of a single training session, or adding assisted breathing training. The course optimization suggestions are pushed to the main interface, and the rehabilitation path is updated after the patient confirms.

5. The interactive method for rehabilitation training of dysphagia after stroke according to claim 1, characterized in that, The psychological assessment sub-interface includes emotion assessment controls and an emotion diary input area; In the psychological assessment sub-interface, an emotion assessment is performed. When an assessment result that matches a preset negative emotion pattern is detected, an emotion regulation mode is triggered, including the following steps: In response to the triggering operation of the emotion assessment control, a standardized depression or anxiety assessment scale is loaded and displayed, the patient's response to the selection of each item option is received, and an emotion score is generated. The patient's subjective description and classification of their emotional state on that day can be received through the text input area and emotion tag selector of the emotion diary input area. Based on the emotion score and the content of the emotion diary, when a combination that matches the preset negative emotion pattern is identified, the emotion regulation mode is triggered, the mindfulness training video playback window is automatically loaded, and emotion regulation suggestions are displayed, prompting the patient to confirm whether to start the intervention process.

6. The interactive method for rehabilitation training of dysphagia after stroke according to claim 1, characterized in that, The AR training sub-interface includes an AR guidance overlay area and gamified feedback controls; In the AR training sub-interface, food or utensils in the field of view are identified in real time, swallowing risk warnings and safe eating guidelines are displayed, and an interactive swallowing game is loaded. Feedback is provided based on the results of swallowing actions, and achievement badges and incentive information are updated, including the following steps: The device's camera captures images of food or tableware within the current field of view, and the recognition results are obtained. Based on the recognition results, a swallowing risk level label and corresponding safe eating operation instructions are generated in the AR guidance overlay area, and displayed as an augmented reality layer overlaid on the real-time image of the food or tableware. In response to the triggering operation of the gamified feedback control, an interactive training task synchronized with the swallowing action is loaded, the swallowing action execution signal is received, real-time visual or audio feedback is output according to the accuracy of the action, and achievement marks and incentive information are updated. If the swallowing integrity score is lower than the preset threshold in two consecutive interactive training tasks, and the patient's emotional label is identified as including "frustration" or "anxiety", the difficulty of the next interactive training task will be automatically reduced.

7. The interactive method for rehabilitation training of dysphagia after stroke according to claim 1, characterized in that, The main interface also includes a rehabilitation progress visualization area, which includes: A line graph showing the changes over time based on the swallowing function scale score; The progress bar for achieving each rehabilitation goal is calculated by weighting the completion rate of the corresponding course and the accuracy of the movements. An achievement graph displays unlocked training levels and incentive badges; The rehabilitation progress visualization area allows patients to initiate course adjustment requests or goal revision requests by clicking on any chart element, thereby re-executing the rehabilitation path generation logic.

8. The interactive method for rehabilitation training of dysphagia after stroke according to claim 1, characterized in that, It also includes steps for collaboration between healthcare professionals: The patient information report interface is displayed on the medical staff's end, and the patient's swallowing function assessment report, rehabilitation path, training completion rate, emotion score and AR training feedback data are synchronized in real time. In response to the instructions from medical staff, suggestions for adjusting dietary plans, corrected training videos, or high-risk warning zones are pushed to the patient's main interface.

9. The interactive method for rehabilitation training of dysphagia after stroke according to claim 1, characterized in that, It also includes steps for family coordination: After obtaining authorization from the patient, grant family members access to view training records, edit diet logs, and synchronize rehabilitation goal dashboards. It receives images of the eating environment uploaded by family members, identifies potential sources of risk, and automatically generates suggestions for environmental modifications.

10. An interactive rehabilitation training system for stroke patients with dysphagia, characterized in that, It includes a patient terminal, a medical staff terminal, a family member terminal, and a cloud server terminal, for executing the interactive method for stroke swallowing disorder rehabilitation training as described in any one of claims 1 to 9; The cloud server is used to enable data communication between the patient, the medical staff, and the family members.

Images