A multimodal evaluation system for teaching of physiotherapy joint mobilization
Patent Information
- Application Number
- CN202610985009.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-03
- Publication Date
- 2026-09-22
AI Technical Summary
[0002]在康复治疗学专业教学中，关节松动术属于典型的高实践性、高规范性技能；学生需要掌握患者体位摆放、治疗师站位、手部接触点、固定手与治疗手分工、操作方向、手法分级、节律、频率、持续时间和治疗后评估等多个环节；传统教学主要依赖教师现场示范和人工评分，评价过程主观性较强，难以对学生每一次训练过程进行量化记录和长期追踪；不同教师之间的评分尺度也可能不一致，导致标准化考核难度较大；现有部分实训系统能够采集力、角度或视频数据，但缺少围绕课程知识点、技能步骤、病例任务和评分指标的完整教学闭环，难以支撑课堂教学、课后练习、OSCE 考核、竞赛训练和教学质控；现有系统通常将传感数据、视频数据、评分数据和学生学习数据分散存储，缺少统一的训练事件模型，难以还原学生从准备、操作、纠错、完成到复盘的全过程；学校在开展康复治疗学实训基地建设和教学改革项目时，需要一种能够体现学校教学成果、课程建设成果和标准化评价能力的系统，而不宜将企业可生产销售的核心实体模拟人结构作为学校专利的主要保护范围
重点体现课程建设、实训考核、过程评价和教学质量分析，适合作为学校教学改革和校企合作成果；不限定具体关节训练装置的机械结构，只规定教学数据、任务模型和评价流程；将体位、手部位置、操作方向、力度、幅度、节律、时间等指标转化为可配置的评分规则，减少教师主观评价差异；通过统一时间轴关联数据、事件、视频和评分结果，可以完整回看学生操作过程；根据学生历史训练数据识别薄弱环节，生成针对性训练建议；学校可基于班级、课程、任务维度分析教学质量，为实训基地建设、课程认证和教学改革提供数据。
Smart Images
Figure CN122799697A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of rehabilitation therapy technology, and more specifically, to a multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy. Background Technology
[0002] In rehabilitation therapy education, joint mobilization is a typical highly practical and standardized skill. Students need to master multiple aspects, including patient positioning, therapist positioning, hand contact points, division of labor between the immobilizing and therapeutic hands, direction of manipulation, grading of techniques, rhythm, frequency, duration, and post-treatment assessment. Traditional teaching mainly relies on on-site demonstrations and manual scoring by teachers, which is highly subjective and makes it difficult to quantify and track each student's training process over the long term. Inconsistent scoring standards among different teachers also make standardized assessment challenging. While some existing training systems can collect force, angle, or video data, they lack a complete teaching loop encompassing course knowledge points, skill steps, case tasks, and scoring indicators, making it difficult to support classroom teaching, after-class exercises, and OSCE (Outcome-Based Education and Testing). Assessment, competition training, and teaching quality control; existing systems typically store sensor data, video data, scoring data, and student learning data in a scattered manner, lacking a unified training event model, making it difficult to recreate the entire process of students from preparation, operation, error correction, completion to review; when schools carry out rehabilitation therapy training base construction and teaching reform projects, they need a system that can reflect the school's teaching achievements, curriculum construction achievements, and standardized evaluation capabilities, rather than using the core physical simulated human structure that can be produced and sold by enterprises as the main scope of protection for school patents.
[0003] Therefore, we have made improvements to this by proposing a multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy. Summary of the Invention
[0004] To address the aforementioned problems in the existing technology, the purpose of this invention is to provide a multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy.
[0005] To solve the above problems, the technical solution adopted by the present invention is as follows: it includes the following modules: The course task configuration module is used to create and configure a teaching task library for upper limb joint mobilization techniques. Each teaching task includes the joint location, patient position, therapist's position, immobilization hand position, treatment hand position, operation direction, manipulation grade, operation range, frequency, time, and scoring weight. The task library is divided into basic cognition, single skills, comprehensive cases, assessment, and competition tasks according to the teaching level, and supports teachers to customize cases and adjust scoring weights. The training terminal access module is used to access the training terminal and receive the multimodal training data output by it. The training terminal includes a bionic training device, a mannequin, a sensor acquisition pad, an angle / pressure acquisition device, an assessment table, a camera, a tablet, or virtual simulation software. This module does not limit the mechanical structure of the terminal, but only specifies the data type and interface format. The multimodal data acquisition module is used to establish a unified timeline for each training session and synchronously collect contact data, posture data, mechanical data, motion data, and video data. Contact data determines the position of the immobilizing and treating hands, posture data determines the patient's position, mechanical and motion data determines the amplitude, rhythm, direction, and grading of the operation, and video data is used for review and tracing. The operation event recognition module is used to convert continuously collected data into standardized operation events, including preparation completion, body position compliance, correct hand contact, stable immobilization of the hand, initiation of treatment, reaching the target direction and amplitude, rhythm stability, operation completion, error occurrence, and teacher correction; the recognition adopts threshold rules, time window rules, step sequence rules, teacher confirmation rules, or AI-assisted rules. The scoring rule engine generates training results based on preset scoring rules. The scoring adopts a combination structure of step score, parameter score, safety score, communication score and debriefing score. The scoring dimensions include training preparation, patient position, hand position, operation direction, technique grading and intensity, rhythm and time, safety and communication. It supports weight configuration, deduction item configuration, veto item configuration and teacher manual correction. Major errors such as failure to judge contraindications, incorrect position, reverse direction, and intensity exceeding the threshold are subject to veto or high deduction. The training process traceability module is used to associate and store training data, operation events and results with a unified timeline, so as to realize the complete playback and traceability of the training process. The teacher-side management module is used for creating course tasks, publishing training plans, editing grading rules, monitoring training, and reviewing grades. The report generation module is used to automatically generate training reports that include scores for each dimension, error steps, data curves, and video indexes.
[0006] Preferably, the multimodal data acquisition module allocates the unified training timeline to different data sources, so that a training process can be completely restored; the playback and tracing of the training process traceability module includes viewing the occurrence time, duration, data curve, video footage, and system judgment results of each key step.
[0007] Preferably, it also includes a student-side training module, used to display training tasks, operation steps, precautions, real-time prompts, error reminders, training results, personalized improvement suggestions, as well as historical training records, score trends, weak points, and recommended practice tasks to students; the student-side training module is deployed on computers, tablets, mobile terminals, virtual simulation terminals, large screens in training rooms, or online teaching platforms.
[0008] Preferably, the system also includes a teaching quality analysis module, which is used to statistically analyze the pass rate, average score, high error rate, and relationship between training sessions and performance improvement in different classes, teachers, and course tasks, and to generate course improvement suggestions. The system exchanges data with the school's academic affairs platform, training room management platform, or virtual simulation teaching platform to output student training results and course achievement status.
[0009] Preferably, the system supports switching between practical training assessment scenarios and classroom practice scenarios: when in a practical training assessment scenario, the assessment mode is enabled, real-time answers are hidden, only the training process is recorded, and a scoring report is generated after the training is completed; when in a classroom practice scenario, the training mode is enabled, errors are prompted in real time, and correction suggestions are provided.
[0010] Preferably, the grade review in the teacher-side management module includes manual review of the system's automatic scoring results, and the manual review retains the scores before and after modification, the reason for modification, the modifying teacher, and the modification time; the training process monitoring includes viewing the training progress, abnormal prompts, and real-time scores of multiple students on the same interface.
[0011] Preferably, the report generation module is also used to automatically generate class statistical reports and course quality analysis reports; the training report includes task name, training time, total score, teacher comments and improvement suggestions, for teaching review and classroom evaluation, and teachers can locate event nodes including incorrect treatment hand position, unstable rhythm, and insufficient amplitude.
[0012] Preferably, in the course task configuration module, the teacher can adjust the parameters of the teaching task according to the curriculum standards.
[0013] Preferably, the course improvement suggestions of the teaching quality analysis module are generated based on the pass rate, average score, high error rate, and relationship between training frequency and performance improvement at the class, course, and task dimensions, providing data support for curriculum reform and practical training base construction.
[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: It focuses on curriculum development, practical training assessment, process evaluation, and teaching quality analysis, making it suitable as a result of school teaching reform and industry-university cooperation. It does not limit the mechanical structure of specific joint training devices, only specifying teaching data, task models, and evaluation processes. It transforms indicators such as body position, hand position, operation direction, force, amplitude, rhythm, and time into configurable scoring rules, reducing subjective differences in teacher evaluation. By linking data, events, videos, and scoring results through a unified timeline, it allows for a complete review of students' operation processes. It identifies weaknesses based on students' historical training data and generates targeted training suggestions. Schools can analyze teaching quality based on class, course, and task dimensions, providing data for practical training base construction, course certification, and teaching reform. Attached Figure Description
[0015] Figure 1 This application provides an architecture diagram of a multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy. Figure 2 The overall architecture diagram of the multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy, and the standardized teaching and assessment system for upper limb joint mobilization techniques, provided in this application; Figure 3 This application provides a diagram showing the relationship between the course task configuration module and the scoring rule engine of a multimodal assessment system for teaching joint mobilization techniques in rehabilitation therapy. Figure 4 This application provides a schematic diagram of multimodal data acquisition and a unified time axis for a multimodal evaluation system used in teaching joint mobilization techniques in rehabilitation therapy. Figure 5 This application provides a flowchart of an operation event recognition system for teaching joint mobilization techniques in rehabilitation therapy. Figure 6 This application provides a multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy, including a training process traceability and video index relationship diagram. Figure 7 This application provides a functional block diagram of the teacher-side management interface of a multimodal assessment system for teaching joint mobilization techniques in rehabilitation therapy; Figure 8 This application provides a student-side training feedback report structure diagram for a multimodal evaluation system used in teaching joint mobilization techniques in rehabilitation therapy; Figure 9 This application provides a schematic diagram of the data flow for a multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy, which includes a teaching quality analysis report. Detailed Implementation
[0016] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.
[0017] Reference Figures 1-9 As shown, a multimodal assessment system for teaching joint mobilization techniques in rehabilitation therapy includes the following modules: The course task configuration module is used to create and configure a teaching task library for upper limb joint mobilization techniques. Each teaching task includes the joint location, patient position, therapist's position, immobilization hand position, treatment hand position, operation direction, manipulation grade, operation range, frequency, time, and scoring weight. The task library is divided into basic cognition, single skills, comprehensive cases, assessment, and competition tasks according to the teaching level, and supports teachers to customize cases and adjust scoring weights. The training terminal access module is used to access the training terminal and receive its output multimodal training data. The training terminal includes a bionic training device, a mannequin, a sensor data acquisition pad, an angle / pressure data acquisition device, an assessment table, a camera, a tablet, or virtual simulation software. This module does not limit the mechanical structure of the terminal, but only specifies the data type and interface format. The multimodal data acquisition module is used to establish a unified timeline for each training session and synchronously collect contact data, posture data, mechanical data, motion data, and video data. Contact data determines the position of the immobilizing and treating hands, posture data determines the patient's position, mechanical and motion data determines the amplitude, rhythm, direction, and grading of the operation, and video data is used for review and tracing. The operation event recognition module is used to convert continuously collected data into standardized operation events, including preparation completion, body position compliance, correct hand contact, stable immobilization of the hand, initiation of treatment, reaching the target direction and amplitude, rhythm stability, operation completion, error occurrence, and teacher correction; the recognition adopts threshold rules, time window rules, step sequence rules, teacher confirmation rules, or AI-assisted rules. The scoring rule engine generates training results based on preset scoring rules. The scoring adopts a combination structure of step score, parameter score, safety score, communication score and debriefing score. The scoring dimensions include training preparation, patient position, hand position, operation direction, technique grading and intensity, rhythm and time, safety and communication. It supports weight configuration, deduction item configuration, veto item configuration and teacher manual correction. Major errors such as failure to judge contraindications, incorrect position, reverse direction, and intensity exceeding the threshold are subject to veto or high deduction. The training process traceability module is used to associate and store training data, operation events and results with a unified timeline, so as to realize the complete playback and traceability of the training process. The teacher-side management module is used for creating course tasks, publishing training plans, editing grading rules, monitoring training, and reviewing grades. The report generation module is used to automatically generate training reports that include scores for each dimension, error steps, data curves, and video indexes.
[0018] Furthermore, the multimodal data acquisition module allocates a unified training timeline to different data sources, enabling a complete reconstruction of a training process; the playback and tracing of the training process traceability module includes viewing the occurrence time, duration, data curves, video footage, and system judgment results of each key step.
[0019] Furthermore, it also includes a student-side training module, which displays training tasks, operation steps, precautions, real-time prompts, error reminders, training results, personalized improvement suggestions, as well as historical training records, score trends, weak points, and recommended practice tasks to students; the student-side training module is deployed on computers, tablets, mobile terminals, virtual simulation terminals, large screens in training rooms, or online teaching platforms.
[0020] Furthermore, it also includes a teaching quality analysis module, which is used to statistically analyze the pass rate, average score, high error rate, and relationship between training sessions and performance improvement in different classes, teachers, and course tasks, and generate course improvement suggestions; the system exchanges data with the school's academic affairs platform, training room management platform, or virtual simulation teaching platform to output student training results and course achievement status.
[0021] Furthermore, the system supports switching between practical training assessment scenarios and classroom practice scenarios: when in a practical training assessment scenario, the assessment mode is enabled, real-time answers are hidden, only the training process is recorded, and a scoring report is generated after the training is completed; when in a classroom practice scenario, the training mode is enabled, errors are prompted in real time, and correction suggestions are provided.
[0022] Furthermore, the grade review in the teacher management module includes manual review of the system's automatic scoring results. The manual review retains the scores before and after modification, the reason for modification, the teacher who made the modification, and the modification time. The training process monitoring includes viewing the training progress, abnormal prompts, and real-time scores of multiple students on the same interface.
[0023] Furthermore, the report generation module is also used to automatically generate class statistical reports and course quality analysis reports; the training report includes task name, training time, total score, teacher comments and improvement suggestions, for teaching review and classroom evaluation, and teachers can locate event nodes including incorrect treatment hand position, unstable rhythm, and insufficient amplitude.
[0024] Furthermore, in the course task configuration module, teachers can adjust the parameters of teaching tasks according to the curriculum standards.
[0025] Furthermore, the course improvement suggestions in the teaching quality analysis module are generated based on the pass rate, average score, high error rate, and relationship between training frequency and performance improvement at the class, course, and task dimensions, providing data support for curriculum reform and practical training base construction.
[0026] Furthermore, the teaching methods and steps are as follows: S1: Teachers create or select upper limb joint mobilization training tasks in the course task configuration module, and configure task parameters, including joint location, case information, patient position, hand contact point, operation direction, grading requirements, and scoring weights. The course task configuration module has a built-in standardized training task library for shoulder, elbow, and wrist joints. Training tasks are divided into basic cognitive tasks, single skill tasks, comprehensive case tasks, assessment tasks, and competition tasks according to teaching levels, and teachers can customize case tasks and adjust scoring weights. S2: When a student enters training or assessment mode, the system assigns a unique training number to the training session and initiates multimodal data collection. When in a practical training assessment scenario, the assessment mode is activated, real-time answers are hidden, only the training process is recorded, and a scoring report is generated after the training is completed. When in a classroom practice scenario, the training mode is activated, errors are prompted in real time, and correction suggestions are provided. S3: The system synchronously collects multimodal training data of students during the operation process through the training terminal. The multimodal training data includes at least hand position, body position, operation parameters, video clips, and step events. The system allocates a unified training timeline to different data sources so that a training process can be completely reproduced. The multimodal training data also includes contact data, posture data, mechanical data, and motion data. Among them, contact data is used to determine whether the fixed hand and the treatment hand are placed in the standard area, posture data is used to determine whether the patient's body position and upper limb placement meet the requirements, and mechanical and motion data are used to determine whether the operation range, rhythm, direction, and grading meet the requirements. S4: The operation event recognition module, based on a preset teaching event model, converts the continuously collected multimodal training data into standardized operation events. These standardized operation events include: preparation completed, body position compliant, correct hand contact, stable hand fixation, initiation of treatment, reaching the target direction, entering the target amplitude, rhythm stabilization, operation completed, error occurrence, and teacher correction. The operation event recognition employs one or more combinations of threshold rules, time window rules, step sequence rules, teacher confirmation rules, or AI-assisted recognition rules. S5: The scoring rule engine calculates scores for each dimension according to the training task and the standardized operation events, and marks the reasons for errors. The preset scoring rules adopt a combined structure, including step score, parameter score, safety score, communication score, and debriefing score. The scoring dimensions include at least training preparation, patient position, hand position, operation direction, technique grading and intensity, rhythm and time, and safety and communication. The scoring rules support weight configuration, deduction item configuration, veto item configuration, and teacher manual correction. For major errors including failure to determine contraindications, incorrect position, reverse direction, and intensity exceeding the safety threshold, a veto or high deduction rule is set. S6: The training process traceability module binds and stores the multimodal training data, the standardized operation events, and the scoring results with the unified training timeline, and supports retrieval by student, class, course, training project, case task, and assessment batch. Teachers can view the occurrence time, duration, data curve, video footage, and system judgment results of each key step of the student during the review. S7: The report generation module generates a student individual training report and a teacher debriefing report based on the scoring results and the multimodal training data. The student individual training report includes the task name, training time, total score, scores for each dimension, error steps, key data curves, video index, teacher comments, and improvement suggestions. The teacher debriefing report is used for classroom review, allowing teachers to pinpoint event nodes such as incorrect treatment hand position, unstable rhythm, and insufficient amplitude. S8: The teaching quality analysis module summarizes data from class, course and task dimensions, and statistically analyzes the pass rate, average score, high error rate and the relationship between training times and performance improvement. It also automatically generates course quality analysis reports and course improvement suggestions. S9: The system exchanges data with the school's academic affairs platform, training room management platform, or virtual simulation teaching platform, and outputs students' training results and course achievement status. Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of protection of this application is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of one or more embodiments of this application as described above, which are not provided in detail for the sake of brevity.
[0027] One or more embodiments in this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of this application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments in this application should be included within the protection scope of this application.
Claims
1. A multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy, characterized in that, Includes the following modules: The course task configuration module is used to create and configure a teaching task library for upper limb joint mobilization techniques. Each teaching task includes the joint location, patient position, therapist's position, immobilization hand position, treatment hand position, operation direction, manipulation grade, operation range, frequency, time, and scoring weight. The task library is divided into basic cognition, single skills, comprehensive cases, assessment, and competition tasks according to the teaching level, and supports teachers to customize cases and adjust scoring weights. The training terminal access module is used to access the training terminal and receive the multimodal training data output by it. The training terminal includes a bionic training device, a mannequin, a sensor acquisition pad, an angle / pressure acquisition device, an assessment table, a camera, a tablet, or virtual simulation software. This module does not limit the mechanical structure of the terminal, but only specifies the data type and interface format. The multimodal data acquisition module is used to establish a unified timeline for each training session and synchronously collect contact data, posture data, mechanical data, motion data, and video data. Contact data determines the position of the immobilizing and treating hands, posture data determines the patient's position, mechanical and motion data determines the amplitude, rhythm, direction, and grading of the operation, and video data is used for review and tracing. The operation event recognition module is used to convert continuously collected data into standardized operation events, including preparation completion, body position compliance, correct hand contact, stable immobilization of the hand, initiation of treatment, reaching the target direction and amplitude, rhythm stability, operation completion, error occurrence, and teacher correction; the recognition adopts threshold rules, time window rules, step sequence rules, teacher confirmation rules, or AI-assisted rules. The scoring rule engine is used to generate training results based on preset scoring rules. The scoring adopts a combination structure of step score, parameter score, safety score, communication score and debriefing score. The scoring dimensions include training preparation, patient position, hand position, operation direction, technique grading and force, rhythm and time, safety and communication. It supports weight configuration, deduction item configuration, veto item configuration and teacher manual correction. Major errors such as failure to judge contraindications, incorrect posture, opposite direction, and force exceeding the threshold are subject to veto or high deduction. The training process traceability module is used to associate and store training data, operation events and results with a unified timeline, so as to realize the complete playback and traceability of the training process. The teacher-side management module is used for creating course tasks, publishing training plans, editing grading rules, monitoring training, and reviewing grades. The report generation module is used to automatically generate training reports that include scores for each dimension, error steps, data curves, and video indexes.
2. The multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy according to claim 1, characterized in that, The multimodal data acquisition module allocates the unified training timeline to different data sources, enabling a complete reconstruction of a training process; the playback and tracing of the training process traceability module includes viewing the occurrence time, duration, data curves, video footage, and system judgment results of each key step.
3. The multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy according to claim 1, characterized in that, It also includes a student-side training module, which displays training tasks, operation steps, precautions, real-time prompts, error reminders, training results, personalized improvement suggestions, as well as historical training records, score trends, weak points, and recommended practice tasks to students; The student-side training module is deployed on computers, tablets, mobile terminals, virtual simulation terminals, large screens in training rooms, or online teaching platforms.
4. A multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy according to claim 1, characterized in that, It also includes a teaching quality analysis module, which is used to statistically analyze the pass rate, average score, high error rate, and relationship between training sessions and performance improvement in different classes, teachers, and course tasks, and generate course improvement suggestions. The system exchanges data with the school's academic affairs platform, training room management platform, or virtual simulation teaching platform to output student training results and course achievement status.
5. A multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy according to claim 1, characterized in that, The system supports switching between practical training assessment scenarios and classroom practice scenarios: when in a practical training assessment scenario, the assessment mode is activated, real-time answers are hidden, only the training process is recorded, and a scoring report is generated after the training is completed; when in a classroom practice scenario, the training mode is activated, errors are prompted in real time, and correction suggestions are provided.
6. A multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy according to claim 1, characterized in that, The grade review in the teacher management module includes manual review of the system's automatic scoring results. The manual review retains the scores before and after modification, the reason for modification, the teacher who made the modification, and the modification time. The training process monitoring includes viewing the training progress, abnormal prompts, and real-time scores of multiple students on the same interface.
7. A multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy according to claim 1, characterized in that, The report generation module is also used to automatically generate class statistical reports and course quality analysis reports; the training report includes task name, training time, total score, teacher comments and improvement suggestions, for teaching review and classroom evaluation, and teachers can locate event nodes including incorrect treatment hand position, unstable rhythm and insufficient amplitude.
8. A multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy according to claim 1, characterized in that, In the course task configuration module, the teacher can adjust the parameters of the teaching task according to the curriculum standards.
9. A multimodal evaluation system for teaching joint mobilization techniques in rehabilitation therapy according to claim 4, characterized in that, The course improvement suggestions in the teaching quality analysis module are generated based on the pass rate, average score, high error rate, and relationship between training frequency and performance improvement at the class, course, and task levels, providing data support for curriculum reform and practical training base construction.